Returns to Acquirers of Privatizing Financial Services Firms: An
International Examination
Kimberly Gleason
James E. McNulty
Anita K. Pennathur*
Department of Finance
College of Business
Florida Atlantic University
777 Glades Rd.
Boca Raton, FL 33431
*corresponding author
pennathu@fau.edu; phone: 954-236-1272.
Returns to Bidders of Privatizing Financial Services Firms: An International
Examination
Abstract
While the literature reports improved performance for privatizing firms, banking markets are
different. Many privatizing financial services firms face unique problems such as an overhang of
problem loans and weak credit cultures and legal systems. We investigate the returns to successful
bidders in privatization acquisitions of financial services firms. We examine both short- and longhorizon performance and test if such acquisitions result in a change in risk for the bidding firm.
Our results show that the cumulative abnormal returns to bidding shareholders are positive,
perhaps reflecting initial optimism that the foreign firm acquiring the privatizing firm would share
in the success associated with privatization. However, we find that this short-run superior
performance is not sustainable for the bidder. Nonetheless, bidders experience an increase in their
total risk following the acquisition. Long-run returns are higher when the target country has an
effective legal system to enforce contracts and protect private property rights.
1
Returns to Bidders of Privatizing Financial Services Firms: An International
Examination
1. Introduction.
We examine the short- and long-run impact on bidders in acquisitions of
privatizing firms in one of the most protected segments of the economy in many countries
– the financial services sector. While there is a substantial body of literature on the
performance of post-privatization entities, and while many studies have investigated US
and cross-border bank acquisitions, our study is the first to investigate short- and longhorizon returns and changes in systematic risk for bidders in financial services
privatizations.
Furthermore, our study provides insights into the returns to both
American and foreign firms bidding on privatizing banks, as well as for both bank and
non-bank acquirers.
The literature suggests that acquisitions of privatizing financial services firms
would present unique opportunities and challenges that have not been examined to date.
In an extensive review of the literature, Megginson and Netter (2001) find that privatizing
non-bank entities improve, sometimes dramatically, in terms of performance and
efficiency following their listings.
However, as Verbrugge, Megginson and Owens
(1999) discuss, banking markets are different for two major, and a number of subsidiary
but closely related, reasons: (1) banking systems play a crucial role in economic growth
in that banks possess unique resources to identify positive net present value projects and
fund them at reasonable cost (e.g. Levine (1997); (2) an efficient banking and financial
system requires a well functioning legal system (e.g., LaPorta et al (1997), but
entrenched political constituencies often represent major obstacles to this objective. In
2
fact, as they note, the state, which regulates the banking system, is often a major source
of opposition to a non-politicized financial system.
In addition, (3) there are efficiency, safety and transparency issues in banking that
are quite complex; (4) reflecting the transparency issue, there is often a significant
problem loan portfolio in the privatized banks, which is partly unrecognized in financial
reports; (5) banking requires unique managerial talent, which is often lacking in former
government-owned banks.
Government owned banks often perform poorly, and poor performance could
make the bank a likely candidate for privatization. For instance, in their study of the
determinants of bank privatization in Argentina, Clarke and Cull (2002) find that poorly
performing banks are more likely to be privatized. Consequently, foreign bidders may be
able to acquire the privatizing entity at a bargain, and also reap the benefits of the
improved performance following restructuring. The barriers to entry in virtually all
banking markets, combined with the forced disclosure inherent in privatization may work
to the advantage of a large foreign bidder, especially when the bidding firm already has
international experience and reputational capital. Governments may be willing to allow
foreign entities in privatization contests to obtain a favorable price for the liberalizing
bank, particularly if development funds from international institutions are at stake. The
entrance into these newly liberalizing markets may provide opportunities to establish an
important first mover advantage.
Furthermore, as anecdotal evidence suggests,
developed country banks, faced with low margins on traditional activities in their home
countries as well as loss of franchise value due to competition from unregulated nonbank
entities, may find it beneficial to pursue higher margin activities abroad.
3
Cross-border privatization acquisitions have implications for the total and
systematic risk of the bidders. The potential decline in systematic risk from international
diversification has been documented in the literature (Errunza and Rosenberg (1982)).
On the other hand, as suggested by Amihud, DeLong, and Saunders (2002), such
acquisitions could increase the insolvency (total) risk of the bidders because of a moral
hazard problem facing regulators. For instance, a US bank may be interested in acquiring
a foreign entity with the knowledge that it could share in the superior performance
following privatization, particularly when one considers the kinds of concessions foreign
governments make in order to induce foreign firms to take over ailing state-owned banks,
such as capital increases, taking over problem loans, and central bank bailout funding1.
If this risky investment fails, the US bank is safe in the knowledge that either the
domestic (US) or foreign regulatory agencies will provide a safety net in the case of
insolvency. Furthermore, Gruben and McComb (2003) suggest that despite short-term
inefficiencies such as overlending and undercharging, newly privatized banks may
engage in super-competitive behavior via short-term efforts to increase market share in
order to obtain long-term profits. Thus, banks may accept current losses if they believe
that the new market share provides them with a positive present value of expected future
returns.
Nonbank bidders, in contrast, do not have similar regulatory constraints in terms
of capitalization and domestic product market diversification as banks, but nor do they
1
An example of the kinds of concessions granted to foreign bidders is the sale of the healthy part of
Agrobanka to GE Capital. The Czech government agreed to pay GE Capital the difference between
liabilities taken over and the value of assets acquired. Vladamir Vana, one of the minority shareholders,
asserted that the transfer was “more than excessive” and confessed to being “puzzled by the Czech National
Bank lending money in any way to a strong foreign partner” (CTK Business News, 1998).
4
enjoy risk subsidization in the form of deposit insurance or too big to fail policies. Their
expansion motivations may be similar to bank bidders – to obtain global market share, to
pre-empt competitors, to diversify, to compete for position within the global network, and
to have a first mover advantage2. Nonbank institutions may have an advantage over bank
bidders from the perspective of host governments. For example, when Newbridge Capital
bid against banking giants JP Morgan and Citibank for a 20% stake in Shenzhen
Development Bank, it was chosen by the selection panel because “it was a non-banking
institution and therefore raised no fears that as a foreign investor it would see a stake in a
Chinese bank as an extension of its existing operations or a Chinese branch”. In other
words, large transnational banks are viewed as more of a threat than nonbank institutions
(Chan, (2002)).
Our results, which are based on an analysis of 97 privatizations, have relevance
for regulators, policymakers, and the international business community.
They indicate
that short-run returns from acquiring privatized financial services firms are positive and
are greater for non-bank bidders (e.g., finance companies such as General Electric
Capital) than bank bidders.
This is in contrast to most studies of domestic bank
acquisitions, which generally find negative returns to acquirers. Nonetheless, long-run
acquirer returns are negative. These long-run returns are positively related to the rule of
law, suggesting again that legal systems affect the performance of financial firms
operating within such systems, along the lines first suggested by LaPorta et al (1997). In
For example,. Metlife acquired Mexican insurer Hidalgo because “the current mix of Hidalgo’s insurance
products plays to Metlife’s core competencies in individual life, group products, and worksite marketing,
and that its product development capabilities can be utilized to update, expand and improve on Hidalgo’s
product mix” (Associated Press, 2002). Despite the optimism, Metlife’s debt was downgraded by Moody’s
immediately following the acquisition.
2
5
addition, the U. S bidders experience a reduction in systematic risk after such
acquisitions.
The remainder of the study is organized as follows: Section 2 describes the
relevant literature. We describe the data and develop our hypotheses and methodology in
Section 3, and present the empirical results in section 4. Section 5 concludes.
2. Literature Review.
Our research draws from extant literature in several areas: (a) privatization; (b)
bank mergers; (c) banking problems in transition economies; and (d) asymmetric
information.
2.1. Privatization and performance
The first strand of literature is research on the impact of privatization on the
operating and financial performance of the firms being privatized. The majority of this
research demonstrates that private ownership leads to better firm performance, and that
foreign ownership, when allowed, is associated with better post privatization
performance.
In general, for both developed and developing economies, the post-
privatization performance is superior. For instance, LaPorta and Lopez-de-Silanes (1999)
report that former Mexican state-owned enterprises (SOEs), while unprofitable prior to
privatization, dramatically improve their performance after privatization, and eventually
become very profitable. Studies examining privatization through share offerings also
document evidence that privatization via public share offerings results in improved
operating and financial performance. Megginson, Nash, and van Randenborgh (1994)
6
and Boardman, Laurin, and Vining (1989) examine such offerings and find significant
increases in post-privatization productivity, output, and efficiency. In a study closely
related to ours, Verbrugge, Megginson, and Owens (1999) document marginally
significant increases in profitability and capital adequacy for commercial banks that are
privatized in OECD countries. For example, return on equity for a sample of banks in
high-information economies increases from 13.7% to 17.1%. They also find that capital
ratios increase, but also report on-going continued partial state ownership in many cases,
which they consider problematic
In their study of acquisitions of former SOEs in transition economies, Uhlenbruck
and Castro (2000) report that SOEs that are privatized via purchase by foreign firms fare
better than any other method of privatization. Such acquisitions require a prolonged and
expensive commitment, and post-privatization performance is related to the strategic fit
between the merging firms, and continued government involvement.
Otchere and Chan (2003) examine the intra-industry effects of the privatization of
the Commonwealth Bank of Australia (CBA).
They find that rival banks reacted
negatively to the privatization announcement. Interestingly, they find that the long-term
performance of CBA improved as the proportion of government ownership decreased
over the three-year post-privatization period. CBA outperformed its peers on financial
and operating ratios in the post-privatization period, leading Otchere and Chan (2003) to
conclude that the best gains in efficiency are obtained from complete privatization.
Along a similar vein, Frydman et al. (1999), and Harper (2002) find that
privatization in transition economies results in better firm performance relative to firms
that remain state controlled. Concentrated private ownership is linked to better firm
7
performance, as is majority ownership by outside investors. Studies such as Black et al
(2000) on the privatization in the former Soviet republics lead to an interesting insight:
privatization through a concentrated insider managerial and ownership structure leads to
poorer post-privatization performance. The best performance is associated with outside
ownership, which, as noted, is the subject of this paper. Djankov (1999) also documents
that foreign share ownership results in considerably improved performance for
privatizations conducted in the former Soviet countries.
Moreover, merely following the correct technical process of privatization is not
sufficient for improving performance. Unal and Navarro (1999) describe the process of
the Mexican bank privatization that occurred in 1991. While the Mexican government
designed a transparent and effective privatization program, they failed to provide
adequate regulatory and legal oversight. The authors suggest that the subsequent banking
crisis of 1994 was a result of this lax oversight.
2.2. Returns to shareholders, changes in risk, and efficiency gains in bank mergers.
Despite voluminous research on domestic bank mergers and acquisitions, there is
relatively little agreement on the impact on shareholder wealth or the efficiency gains
from such mergers. For instance, while Hannan and Wolken (1989) and Houston and
Ryngaert (1994) find no wealth gains from mergers, Cornett and Tehranian (1992) and
DeLong (2003) find positive shareholder wealth effects.3 Along the same lines, while
Spindt and Tarhan (1993) find that operating gains accrue due to scale economies in
mergers of small banks, other studies (for e.g., Berger and Humphrey (1992), Rhoades
3
In DeLong (2003) the positive returns were only experienced by banks in which the mergers were well
focused.
8
(1993), and Pilloff (1996)) do not find any operating gains from bank mergers. The
potential gains from mergers can be even more elusive when banks attempt to merge
across national lines. Although managers no doubt seek and anticipate gains from such
mergers, it appears that cultural and organizational differences make long-term gains
difficult to realize.
Amihud, De Long, and Saunders (2002) examine the impact of cross-border bank
mergers on the risk and abnormal returns of acquiring banks. They find that while the
acquirer’s risk does not change, the abnormal returns to the acquirer are significantly
negative. Focarelli and Pozzolo (2001) suggest that asymmetric information and
regulation reduce the number of cross border acquisitions. Moreover, they conclude that
the banks that expand abroad tend to be the large and efficient banks that are located in
developed banking markets. Confirming these results is the study by Buch and DeLong
(2003) who examine over 3000 cross-border bank mergers that occur between 1985 and
2001. They consider such mergers relatively rare and find support for the notion that one
reason for this is the high information costs associated with acquiring a bank in another
country. Especially relevant to our study is their finding that a very common occurrence
is a bank from a developed country acquiring a bank from a developing country.
Information costs are important variables in determining whether a bank merger occurs
between a country pair. Acquisitions are more likely when two countries have the same
language, the same legal system and are not far from one another.
9
2.3. Banking Problems Arising From State Ownership.
A well developed literature exists on banking and overall economic problems in
the former communist countries as a result of their transition to a market economy; this
literature provides examples of the types of problems that foreign financial service firms
may face when they acquire privatizing financial institutions. (We explain the relevance
of this literature to the other sample countries below (see Section 2.3.4). In addition,
approximately 25% of the privatizations analyzed in this paper occur in these countries.
2.3.1 The Economic Environment.
La Porta, de Silanes, and Shleifer (2002) provide an overview of government
ownership of banks around the world. Their comprehensive research indicates that even
in countries which are relatively developed and market oriented, such as Australia,
Germany, and Austria, government ownership is nontrivial, and increasing levels of
government involvement in the banking sector are related to lower levels of economic
growth. The textbook case of the sundry inefficiencies produced by state ownership is
the transition economies of Central and Eastern Europe. Campos and Coricelli (2002)
note that expectations at the beginning of the transition in the early 1990’s were
extremely positive because of the move toward liberalized markets. However, the actual
experience has been quite different. Output fell in each country during the first four years
and in many the decline continued.
2.3.2. Credit Markets and Problem Loans.
One of the problems foreign banks face in entering many of the transition
economies is weak credit standards and an enormous volume of bad loans. Newly
10
created commercial banks opened their doors in the 1990s with a large volume of bad
loans inherited from central planning, and even new loans quickly became problem loans
((Buch (1996) p.29). Scholtens (2000) reports very high levels of problem loans/total
assets in many of these countries. Banks, however, often have an incentive to continue to
fund weak or failing enterprises because to do otherwise would create large loan writeoffs and cause the firm to contract operations and employment ((Perotti (1993), Gorton
and Winton (2001)). There is a striking similarity to the U.S. savings and loan crisis
since the undermining of credit standards and the weak competitive and regulatory
environment caused many of the new banks to fail (Gorton and Winton (1998, p. 634)).
2.3.3 Legal Systems.
In many of the transitional countries, banks are reported to be unwilling to lend to
small firms because contracts are either unenforceable or very difficult to enforce.4
Weaknesses in legal systems mirror weaknesses in other institutions that are normally
part of a market economy. Developing a banking system and a credit culture without the
rule of law to protect private property transactions is a difficult problem. Rapaczynski
(1996) argues that this problem cannot be corrected quickly because legal systems are
endogenous, and property rights are quite complex. Pistor Raiser and Gelfer (2001)
document significant problems in enforcement of business contracts in some of the
transition economies, including Russia and Ukraine (both of which are in our sample).
4
This problem is discussed in detail in the various studies in Entrepreneurship (1997), some which have
been published in Wachtel (1999).
11
2.3.4. Relevance of this literature to other sample countries.
Similar problems exist in other countries in our sample. For example, Baer and
Nazmi (2000) describe how decades of high inflation in Brazil distorted bank decisionmaking. Just like the former Soviet countries, Brazilian banks were faced with large
volumes of problem loans when the monetary authorities belatedly attempted to bring
inflation under control. The need for new revenues caused lenders to extend credit to
riskier borrowers, which further aggravated the bad loan problem. Also, just like the
former communist countries, lenders in this environment were frequently unable to
distinguish good loans from bad loans, which may have led to a restriction of credit
because of an adverse selection process. Similar overlending experiences have been
reported for the Brazil and the Philippines by Hassan (2002), and also for a number of
other sample countries in various conference volumes (e.g., Caprio et al, (1998),
Kaufman (1999)). Corruption is also prevalent in countries where the rule of law is not
strongly established; several firms in our sample were subsequently investigated for
illegal or unethical business transactions during the bidding process.5
Smith and Malkin (2001) consider the Mexican government bailout during the
peso crisis of 1994 and subsequent sale of banks to foreign buyers. All did not work as
planned for the shareholders of foreign banks entering these markets; apparently the
uncompetitive insider “sweetheart” deals were not all they were touted to be and the
5
An example is Spanish BBV, whose executives admitted that they used cash hidden offshore in slush
funds to convince a Mexican official to allow them to acquire a stake in Probursa. BBV confessed that it
reported its $239 million secret accounts as extraordinary gains during the year (Brooks, 2003). BBV was
also implicated in a bribery scandal involving Alberto Fujimori and his security advisor in the 1995
Peruvian Banco Continental acquisition.
12
challenges of restructuring were understated. In the case of the acquisition of Conifa by
Citibank, Smith and Malkin (2001) point out: “Even though the bank had been cleaned
up by Mexican banking regulators, it was still a mess…and it took several years for
Citibank to integrate its operations with Conifa.”
Lane (2003) finds that macroeconomic volatility is approximately twice as great
in emerging markets as in developed countries. Thus, the abovementioned problems,
many of which resulted from such volatility in, e.g., the transition economies, should be
much less in the developed countries in our sample (e.g., Austria, Italy and New
Zealand). On the other hand, the emerging markets may provide diversification benefits
for acquiring banks. As described by Castillo (1997), when explaining why European
banks tended to dominate the acquisition of privatizing Argentine banks, “European
banks like to do business in Mercosur countries where their investments tally with their
other strategic interests. Large transnational banks feel the need to diversify their
holdings, taking numerous positions in developing countries in order to minimize their
overall risks in a famously volatile sector.”
2.4. Asymmetric Information.
Boot (2002) concludes a detailed review of the literature on bank relationships
with the observation that the reduction in asymmetric information is the major function of
banks and their major source of value creation.
Relationships with borrowers are
probably more problematic and asymmetric information may be more difficult to reduce
when financial services firms acquire banks in other countries. In the context of the
present paper, there is an additional asymmetric information problem between the
13
managers of the acquiring bank and the loan officers at the acquired bank. The two
groups do not share a common corporate culture, are not used to doing business the same
way and often have different languages and legal systems6. Loans that appear to be good
loans to one group may appear to be potential problem loans to the other group, and vice
versa. On the other hand, given Boot’s observation, banks in which the two management
groups work together effectively to resolve these problems should have major
opportunities to create value for shareholders. In addition, banks acquiring privatized
banks in transition and other developed countries have a wealth of skills in credit analysis
loan monitoring that is sorely needed in these countries. Thus, asymmetric information
issues provide one reason why we might expect to see positive announcement period
returns from such acquisitions.
3. Data and methodology
3.1 Data.
We obtain announcements of acquisitions of privatizing financial services firms
from the Securities Data Corporation (SDC) International Mergers and Acquisitions
database and from World Bank Privatization database.
Specifically, we search for
announcements via a keyword search using target Standardized Industrial Classification
(SIC) code in the 6000s, privatization flag = yes, and acquirer ultimate parent stock
exchange as the New York, NASDAQ, or American Stock Exchange. We also use target
ultimate parent public status = government in the SDC database.
SDC
provides
Castilho (1997), using the example of Argentine bank acquisitions, elaborates “multinational banks bring
in new banking technology and manageable long term credit mechanisms… providing skills that were
largely lost during the Latin American banks’ long years enduring high inflation, an experience that left the
region’s bankers with the habit of putting trust in short term credit first.”
6
14
detailed information on both the acquirer and the target, and the sample from the World
Bank database lists all privatization transactions as well as entity size, form of
privatization (voucher, listing, or sale to a domestic or foreign investor) and SIC code.
Our initial sample comprises 121 privatizing financial services firms.
We then verify these announcements with a Lexis Nexus search, which reduces
our sample to 105 verifiable announcements. We also select financial services SIC codes
and search Lexis Nexis for news regarding privatizing financial services entities that were
listed as sold to a foreign entity. We obtain financial data on the US-listed acquirers from
Standard and Poor’s Research Insights database. Target data are obtained from Banker’s
Almanac, news announcements surrounding the privatization, SDC, and the World Bank.
Stock price data for successful bidders are obtained from the Center for Research in
Securities Prices (CRSP) tapes. Acquiring firms which do not have prior year Research
Insights data, or CRSP data 110 days prior to and following the announcement date are
removed from the sample. The final sample of 97 transactions comprises only US and
foreign bidders for which data are available from CRSP, producing a unique sample
consisting of US and foreign bidders that acquire privatizing banks and nonblank
financial services firms in both culturally related and culturally dissimilar areas. We also
set up a control sample of similar non-acquiring financial services firms from the same
SIC codes and size groups as the sample firms.
We use various macroeconomic data for the target country to help us understand
the determinants of both short and long run returns. Nominal GDP (1986-1997) is from
the United Nations Statistical Yearbook CD-ROM and World Bank Development
Indicators Online (1997-2002). Since the two sources report different results, the series
15
are linked at 1997, the last year that the UN data is available. GDPit/GDPUSt is Gross
Domestic Product in current U.S. dollars for the year of the acquisition for the country
where the privatized bank is located divided by nominal U.S. GDP for the same year.
This is a measure of market size. We hypothesize that both short and long run returns
will be higher if the market served by the target bank is larger.
We use a relative measure because the acquisitions take place over a long time
period (1986 to 2002). It is difficult to directly compare absolute dollar GDP figures
from different countries for different years. For example, if GDP in Brazil were five
percent of U.S. GDP in 1986 and five percent in 2002, then Brazil would be
(approximately) as attractive a market to acquirers in 2002 as it was 16 years earlier. 7
We use the same procedure (and sources) for per-capita GDP (PCIit/PCIUSt).
Rule of Law is a measure of the extent to which private property rights and
contracts are governed and protected by well functioning legal systems.
This is a
commercial measure published by Political Risk Services (prsgroup.com) entitled “Law
and Order” and is the same data series used by LaPorta et al (1997) in their classic study
of the impact of legal systems on financial markets. Each country is rated on a scale of
one to six, with six reflecting the greatest protection of private property rights and
enforcement of contracts.
< insert Table 1 about here >
7
The alternative of using only GDPit is rejected because it implies that a market is more attractive
in, say 2002 than in say, 1986 simply because nominal GDP has risen. However, the increase could have
been anemic, or due to inflation, or both. Our procedure also implicitly adjusts for inflation. For example,
if inflation (and real growth) were precisely the same in Brazil and the U.S. between 1986 and 2002, ceteris
paribus, Brazil would be just as attractive a market in both years. If the increase in GDP were due
primarily to inflation, both the exchange rate and GDP it/GDPUSt would decline.
16
The number of acquisitions for the different time periods is presented in Panel A
of Table 1.
Table 1A indicates that the announcements are distributed over time,
although the largest number of transactions occurred between 1995 and 1998. The
distribution of the sample by bank/nonbank, US/foreign, by percent acquired, and by
geographic distribution is shown in Table 1B. While 56.8% of the sample consists of
depository institutions – SIC codes 6021 and 6022 – many of the bidders (43.3%) are
nonbank financial services institutions8.
This number is not surprising, given that
nonbank institutions have scaled up their competitive engagements with banks
domestically. About 48% of the sample consists of banks headquartered in the US, while
52% are foreign banks with equity traded in the US. Foreign banks, such as Banco
Santander, and foreign financial services firms, such as Aegon, facing growth limitations
at home, may anticipate less competition and greater growth opportunities in markets
with which they have a larger cultural familiarity relative to US banks. Panel B of Table
1 also provides detail on the ownership impact resulting from such acquisitions.
Approximately 32% of sample transactions involve an equity stake of 100%.
Table 1 Panel C provides information regarding the distribution of the sample by
location.
About 25% of the sample targets are in the transitional economies;
approximately 40% are located in Southeast Asia and Latin America. The countries with
the largest number of transactions (not shown in the table) are China, Hungary, Poland
and Mexico. Developing countries were not the only ones engaged in financial services
liberalization during the time period covered; many western European countries also
allowed foreign firms to bid for firms that had previously been state owned, including
Italy, Australia, and New Zealand.
8
As noted, an example of a sample nonbank bidder is General Electric Capital.
17
< insert Table 2 about here >
Table 2 provides descriptive statistics for bidding sample firms as well as their
control counterparts. Table 2 indicates that the sample firms are extremely large in terms
of both assets and market value. Mean assets are $119.8 billion; mean market value is
$33.08 billion9. These averages indicate that, for the most part, only the largest entities,
are able to participate in privatization. Acquiring firms are larger than the control group
in terms of both asset size and market value. In addition, the sample firms are more
profitable than their matched firms in terms of return on equity; while their return on
assets and net profit margins are insignificantly different.
For the subsample of bidders that are financial services firms, we examine
indicators of bank profitability. Sample banks, with a loan to asset ratio of 42.7%, are
less dependent on traditional sources of banking revenue than matched financial services
firms. They have lower charge-offs, lower provisions for loan losses, but a higher level
of nonperforming assets, in the year prior to the acquisition.
3.2 Methodology.
3.2.1. Event Study.
Event study methodology is used to identify the wealth effects to bidders and
targets associated with announcements of acquisitions of privatizing financial services
firms. The ordinary least squares market model is used to specify the returns generating
process. Daily excess returns (ARs) are computed by estimating the market model
parameters over the period from t = -110 to t = -11 relative to the announcement day t=0.
18
The standardized cross-sectional method (Boehmer, Musumeci, and Poulsen (1991)) with
Scholes-Williams (1977) betas is used to test for significance. The average excess return
for each day is calculated by summing the ARs for the N firms in the sample and dividing
by N. The cumulative average excess returns (CARs) over a multi-day event period are
calculated by summing the average excess returns over the t day event window.
3.2.2. Long-Horizon Returns.
We hypothesize that acquirers will experience long run losses from acquisitions if
the costs of restructuring outweigh the benefits of any favorable terms and early entrant
advantage. Using the Barber and Lyon (1997) procedure, we estimate matching sampleadjusted long-horizon average holding-period abnormal returns (AHAR) for our sample
bidders. AHARs are calculated starting with the month after the announcement date for
6, 12, and 18 month holding periods.
3.2.3. Changes in accounting performance.
We hypothesize that bidders of privatizing financial services firms may
experience gains or losses in terms of both traditional accounting measures as well as
performance of their loan portfolios. We examine performance changes using controladjusted univariate tests.
9
For comparison purposes, the mean assets of banks with data in S&P Research Insights between 1992 and
1998 was $5.28 billion. Thus, bidders in privatization acquisitions were, on average, 35 times larger.
19
3.2.4. Changes in Risk.
We hypothesize that the acquisition of a privatizing financial services firm may be
a way for bidders to lower systematic risk, relative to the home market, by obtaining the
diversification benefits of operations in markets that are, for the most part, not fully
integrated into the global economy. For each bidding firm, we estimate the changes in its
risk characteristics surrounding the announcement of the acquisition.
The pre-
announcement risk measure, VARPRE [post-announcement risk measure, VARPOST ] is
estimated over the period from t-110 to t-11 [t+11 to t+110], where t=0 is the
announcement date for both sample firms and matched firms. We estimate (a) total risk,
or Var (Ri ), where Ri is the security return on the ith security, and (b) systematic risk, or
i.
4. Empirical Results.
4.1. Short-horizon Returns.
< insert Table 3 about here >
Table 3 presents mean and median cumulative abnormal returns (CARs) and the
number of cumulative abnormal returns that are positive and negative for the bidders and
targets for two-day windows (-1,0), event date (0,0), and three day windows where day 0
is the announcement day. Panel A shows that the announcement of the acquisition is a
wealth generating event for the bidders in general; the three-day event window CAR is
0.61%, which is significant at the 5% level. We further parse the bidders by US or
foreign and participant type and present these results in Panel B. While US bidders
20
experience significant positive CARs, interestingly, we find that the largest and most
statistically significant gains (0.86%) at the announcement accrue to US nonbank bidders.
US banks do not experience significant CARs at the privatization announcement.
Similarly, all foreign bidders experience a positive three-day event window CAR of
0.53%; further parsing demonstrates that it is the foreign non-bank bidders that
experience the significant gains. Thus, unlike Amihud, DeLong, and Saunders (2002)
who find negative abnormal returns for announcements of cross-border acquisitions,
financial services firms that acquire other such firms through privatization are met with a
positive market reaction.
We suggest that the market perceives some of the
abovementioned benefits from privatizing as accruing to the firm that acquires the
privatizing entity. These results also are contrary to other studies on the effects of
domestic (US) bank mergers (e.g., Houston and Ryngaert (1994)) and Hawamini and
Swary (1990)).
< insert Table 4 about here >
We attempt to explain the returns summarized in Table 3 by regressing them on
several explanatory variables. We focus on the (-1,0) abnormal returns. To evaluate
robustness, we estimate seven separate models and present the results of these crosssectional regressions in Table 4. Event window abnormal returns are less for larger
bidders (Log (Market Value)) and for US banks.
Furthermore, participant type
significantly impacts the event window abnormal return; consistent with the raw data in
Table 3, CARs are significantly lower when the bidder is a bank as opposed to a nonbank
21
firm. We suggest that this negative relationship could be due to market concern about
regulation constraining bank bidders. Concern about excess interference in the operation
of a privatized banking firm or restrictions on their activities may lead the market to place
a lower valuation on acquisitions in which the bidding firm is a bank. We also find that
CARs are negatively impacted when bidders acquire a majority stake in the privatizing
firm. (This effect is less pronounced than the others since it is only significant in three of
the seven regressions and only at the 10% level.) Interestingly, returns are also less when
the privatization and acquisition is conducted in a transition economy.
We also test if the event window returns are related to changes in acquirer total
risk and/or systematic risk. Similar to Ahimud, Delong, and Saunders (2002), we find
that event window CARs are positively related to the change in the total risk of the
bidder.
Assuming, for the sake of argument, that market participants expect these
changes in risk, one possible explanation is that investors in the acquiring bank are exante confident in the regulatory umbrella and safety net and are willing to pay higher
share prices even when they expect the total risk of the bidding firm to increase. Our
cross-sectional analysis of short-run returns indicates that factors such as the rule of law
in the host country, GDP, per capita income, and change in the systematic risk measure
are not important considerations to market participants evaluating these acquisitions.
22
4.2. Changes in Risk.
Table 5 provides details on changes in risk for the acquiring banks after the
merger announcement. We find that total risk increases for the sample of all bidders.
Subsampling by US vs. foreign firms shows that the total risk increase is statistically
significant only for US bidders. In addition, we find that total risk increases for all banks,
although only at the ten percent significance level, with US banks driving this result.
Furthermore, nonbank bidders also experience a significant increase in their total risk,
(significant at the five percent level) and once again, the US non-bank bidders drive this
result. Taken together, the implications of these findings are as follows. US bidders in
general increase their total risk in the privatization process. However, the regulatory
impact is different for banks and for nonbanks. As mentioned earlier, acquiring banks
have more incentive to take on risky lending if they are confident in the regulatory safety
net. As the acquiring banks in our sample increase their total risk, the burden of the risk
taking behavior shifts from the banks to the regulatory agencies, which is the classic
moral hazard problem.
In addition, we find from Table 5 that all US bidders and US bank bidders in
particular reduce their systematic risk exposure with the privatization process. Perhaps
banks, due to more stringent regulatory oversight than nonbank entities, are more
concerned with risk-adjusted return improvements and seek to engage in acquisitions that
provide diversification benefits by exploiting market segmentation. In contrast, Amihud,
DeLong, and Saunders (2002) do not find that cross-border bank mergers result in a
change in either total or systematic risk.
< insert Table 5 about here >
23
4.3 Long-Horizon Returns.
These initial positive market reactions are not sustained since the long-term
holding period returns are negative. We present these results for six months, twelve
months, and eighteen months after the acquisition in Table 6. For the overall sample,
long-horizon average holding period eighteen-month returns are negative. Again parsing
the sample by US vs. foreign bidders, we find that US bidders experience negative
returns in all three periods tested, with statistically significantly lower long-run returns in
the eighteen months following the acquisition. While foreign bidders initially experience
statistically significant positive six-month holding period returns, they are unable to
sustain this momentum. Further sub-sampling indicates that these six-month gains accrue
primarily to non-bank foreign bidders. Thus, while the market is initially optimistic
about foreign firms conducting privatization acquisitions, perhaps because it expects
synergy due to cultural similarity, further long-horizon average holding periods are not
positive. Moreover, bank bidders fare significantly worse in both the twelve-month and
eighteen-month horizons than do non-bank bidders.
< insert Table 6 about here >
We provide an analysis of the potential causes of the poor long-horizon postprivatization performance in Table 7 by examining percentage changes in several key
operating performance ratios for the bidders as compared to the abovementioned control
sample. We calculate the change from three years prior to three years post-privatization.
< insert Table 7 about here >
24
Profitability, as measured by return on equity (ROE) and the net profit margin
(NPM), deteriorates for the acquirers over this period; this decline is significantly
different from the control group, which actually experiences an increase. The percentage
change in operating expense, SGAX/Total Assets (the ratio of commercial expenses of
operation incurred in the regular course of business pertaining to the securing of
operating income (but not allocated to costs of goods sold) to total assets) is not
significantly different for the bidding firms relative to the control firms. Similarly, the
change in the loans to total assets ratio is not significantly different for the bidders and
the control firms. Finally, the change in both the non-performing assets/total assets ratio
and the net charge offs to net income ratio is quite different for the two samples of firms.
Both non-performing loans and charge-offs increase for the acquirers and decrease for the
control group. Thus, the change in operating performance is worse for the acquirers in
several areas compared to their non-acquiring peers.
<insert Table 8 about here>
We next investigate the factors that affect longer-term performance with
multivariate regressions with 12-month control adjusted buy and hold returns as the
dependent variable. The regression models in Table 8 indicate that the most important
determinants of the twelve month long-horizon returns are: (a) the size of the bidder (Ln
(Market Value)) which has a positive impact; (b) the rule of law measure, which is also
positive; and (c) the bank/non-bank status of the bidder (banks have lower returns); and
(d) the ROE of the bidder, which surprisingly has a negative impact. The positive
25
relationship to the rule of law measure indicates, not at all surprisingly, that when a US
(or foreign bank listed in the US) acquires a privatized bank in a country which has a well
developed system of protecting private property rights and enforcing contracts, it
experiences better long-term returns.10 This result is quite robust. Similarly, long-term
performance is better for larger bidders, perhaps because the market views them as better
able to manage political risk and operate successfully in different business cultures. An
example is Citibank, which has long-term experience operating in about 100 countries.
Performance is negatively impacted when the acquirer is a bank. Perhaps regulatory
requirements in the home country reduce the ability of the bank to make the necessary
investment to achieve maximum performance from the target. This result is consistent
with the regressions on event date CARs; the market perceives that non-banks would
generate more wealth than banks upon the announcement of the acquisition.
Majority acquisition and whether the bidder is a US firm or not are insignificantly
related to long run returns.
Interestingly, we find that long-term performance is
negatively related to ROE. We surmise that better performing firms are able to obtain
higher returns on contributed capital domestically, and when they expand overseas, they
may overextend limited resources. Firms facing growth constraints and limited returns
on equity at home may find it advantageous to expand in foreign markets, especially
those where their services and products are relatively new.
10
We also estimate the regression using a corruption index, and do not find that long-horizon performance
is impacted by target country corruption factors. To evaluate the effect of asymmetric information, we
also used cultural similarity variables (e.g., common language, geographic proximity) similar to those used
by Buch and DeLong (2003) and do not find any significant relationship.
26
5. Conclusions
We analyze the performance of bidders of privatizing financial services firms.
Our interest in this subject stems in part from some research on domestic and
international mergers that reports that acquisitions can be wealth destroying events for
bidding banks, and from research that demonstrates that privatizations result in
significantly improved firm-level performance. Perhaps acquirers of privatized firms can
share in this improved performance. We find that, unlike the findings of prior research,
acquisitions of privatizing financial services firms are wealth-generating events to the
bidders at the time of announcement. In particular, non-bank firms that perform such
acquisitions experience significantly positive wealth gains. We find evidence that event
window returns are less when the bidders are banks, perhaps signifying some market
concern about continued regulation of a privatizing entity.
While the initial reaction to the privatization announcement is positive, our results
show that these returns are not sustained; in other words, the long-horizon returns are
negative for the bidders. Thus, the costs of assimilating privatizing firms may be greater
than any benefits. We perform some further analysis along these lines by examining
several key ratios for the bidders and a control group of financial services firms, prior to
and post-privatization.
This analysis shows that the bidders fare worse than their
counterparts on changes in ratios such as the net profit margin and charge-offs. Overall,
although there is an initial market enthusiasm for bidders acquiring privatizing firms, the
long-term performance for bidders is weaker than the control group of non-acquirers. We
also find that such acquisitions lead to an increase in total risk for bidding firms;
27
however, for US banks, lower returns are commensurate with lower systematic risk
relative to the US market.
Our study is relevant to regulators, policymakers and for the international
financial services community. Of particular interest is the robust positive relationship of
long run returns to the rule of law measure. As we would expect, when a bidder acquires
a privatized bank in a country which has a well developed system of protecting private
property rights and enforcing contracts, it experiences better long-term returns than other
banks in the sample. This suggests that in many cases, such as acquisitions in some of
the transitional economies of Central and Eastern Europe, the market did not anticipate
that elements of the infrastructure of a market economy, including legal protections and
other cultural factors taken for granted in the home country, are severely underdeveloped
and may remain so for some time. Also of interest is our finding that the bank bidders
are able to reduce their systematic risk, relative to the home country market, following
such acquisitions.
From a regulatory perspective, the increase in total risk after
privatization acquisitions, and the relationship between event-window returns and change
in total risk, indicate both an increase in risk-taking behavior by bidding banks and an
implied confidence in the regulatory safety net.
28
References
Amihud, Y., DeLong, G., Saunders, A., 2002, The effects of cross-border bank
mergers on bank risk and value, Journal of International Money and Finance 21,
857-877.
Baer, W., Nader, N., 2000, Privatization and restructuring of banks in Brazil, Quarterly
Review of Economics and Finance, 40, pp. 2-24.
Barber, B., Lyon, J., 1997, Detecting long-run abnormal returns: The empirical power
and specification of test statistics, Journal of Financial Economics 43, 341-372.
Berger, A.H., Humphrey, D.B., 1992, Megamergers in banking and the use of cost
efficiency as an antitrust defense, Antitrust Bulletin 37, 541-600.
Black, B., Kraakman, R., Tarassova, A., 2000, “Russian privatization and corporate
governance: What Went Wrong? Stanford Law Review.
Boardman, A., Vining, A.R., 1989, Ownership and performance in competitive
environments: A comparison of the performance of private, mixed, and stateowned enterprises, Journal of Law and Economics 32, 1-33.
Boehmer, E., Musumeci, J., Poulsen, A.B., 1991, Event-study methodology under
conditions of event-induced variance, Journal of Financial Economics 30, 253272.
Boot, Arnoud W. A., 2000, Relationship banking: What do we know? Journal of
Financial Intermediation 9, 7-25.
Brooks, Bradley. 2003, Spanish bank admits to secret account.
Associated Press, Rio De Janeiro, Brazil (January 17).
Buch, C., 1996, Creating Efficient Banking Systems: Theory and Evidence from Eastern
Europe, (J.C.B. Mohr, Tubingen).
Buch, C., DeLong, G., 2003, Cross border bank mergers: What lures the rare
animal, Journal of Banking and Finance (Forthcoming).
Campos, N. F., Coricelli, F., 2002, Growth in transition: What we know, what
we don’t and what we should, Journal of Economic Literature 40 (3), 793-836.
29
Caprio, Gerald, William C. Hunter, George G. Kaufman and Danny M Leipziger, 1998,
Preventing Banking Crises: Lessons From Recent Global Bank Failures,
Washington D.C.: World Bank.
Castilho, Carlos. June 8, 1997, Latin America: Europeans scooping up region's privatized
banks, InterPress Service.
Chan, Elaine. 2002. US control set at Shenzhen Bank, WorldSources Emerging Market
Datafile, Hong Kong Imail (October 17).
Czech News Agency, 1998, CNB pays KC20BN to raise equity of GE Capital Bank,
CTK Business News Wire (July 21).
Clarke, G.R.G., Cull, R., 2002, Political and economic determinants of the
likelihood of privatizing Argentine public banks, Journal of Law and
Economics 45, 165-198.
Cornett, M.M., Tehranian, H., 1992, Changes in corporate performance associated
with bank acquisitions, Journal of Financial Economics 31, 211-234.
DeLong, G., 2003, Does long-term performance of mergers match market
expectations? Evidence from the US banking industry,” Financial Management
32, 5-25.
Djankov, S., 1999, Ownership structure and enterprise restructuring in six newly
independent states, Comparative Economic Studies, 41;1, 75-95.
Entrepreneurship in the Transition Economies of Central and Eastern Europe, 1997,
Conference Proceedings, New York University, Stern School of Business
(November).
Errunza, V., and B. Rosenberg, 1982, Investment in developed and less developed
countries, Journal of Financial and Quantitative Analysis 17, 741-762.
Focarelli and Pozzolo,
Frydman, R., Gray, C.W., Hessel, M., Rapaczynksi, A., 1999, When does privatization
work? The impact of private ownership on corporate performance in transition
economies, Quarterly Journal of Economics 114, 1153-1191.
Gazeta Wyborcza, 1999. Allied Irish Banks - Re Bank Zachodni. Polish Daily
Newspaper (April 14).
Gorton, G. and Winton, A., 1998, Banking in transition economies: Does efficiency
require instability? Journal of Money, Credit and Banking 30 , 621-650.
30
Gruben, W.C., and McComb, R.P., 2003, Privatization, competition, and
supercompetition in the Mexican commercial banking system, Journal of
Banking and Finance 27(5) 229-249.
Hannan, T.H., and Wolken, J.D., 1989, Returns to bidders and targets in the acquistion
process: Evidence from the banking industry, Journal of Financial Services
Research 3, 5-16.
Harper, J. T., 2002, The performance of privatized firms in the Czech Republic,
Journal of Banking and Finance 26, 621-649.
Hassan, Mohamed, 2002, Three essays on capital flows, banking weakness and real
exchange rates in East Asia and Latin America, Unpublished doctoral dissertation.
Hawawini, G., and Swary, I., 1990, Mergers and Acquisitions in the US Banking
Industry, North Holland, Amsterdam.
Houston, J.F., Ryngaert, M.D., 1994., The overall gains from large bank mergers,
Journal of Banking and Finance 18, 1155-1176.
Kaufman, G. G., 1999, Bank problems: A global perspective, Research in
Financial Services and Public Policy, 11, Stamford, CT: JAI Press.
Levine, R., 1997. Financial development and economic growth: Views and agenda,
Journal of Economic Literature 35 (June), pp. 688-726.
Lane, Phillip, 2003, Business cycles and macroeconomic policy in emerging market
economies, International Finance, 6 (1), pp. 89-108.
La Porta, R., Lopez-de-Silanes, F., Shleifer, A., Vishny, R., 1997. Legal determinants of
external finance, Journal of Finance 52, (July), pp. 1131-1150.
La Porta, R. and Lopez-de-Silanes, F., 1999, Benefits of privatization – evidence from
Mexico, Quarterly Journal of Economics 114,1193-1242.
La Porta, R., Lopez-de-Silanes, F., and Shleifer, 2000, Government ownership of
banks”. Harvard University Working Paper.
Megginson, W.L., Nash, R.C., and van Randenborgh, M., 1994, The financial and
operating performance of newly privatized firms: An international empirical
analysis, Journal of Finance 49, 403-452.
Megginson, W. L. and Netter, J., 2001, From state to market: A survey of empirical
studies on privatization, Journal of Economic Literature 35, 2 (April), 321-389.
31
Otchere, I., and Chan., J., 2003, Intra-industry effects of bank privatization: A clinical
analysis of the privatization of the Commonwealth Bank of Australia, Journal
of Banking and Finance, 27(5), 949-975.
Park, Zeno, 1999, Seoul shelves sale of top banks amid corporate restructuring analysis.
Associated Press, Seoul. (September 14).
Perotti, 1993, Bank lending in transition economies, Journal of Banking and Finance 17,
1021-1032.
Pilloff, S.J., 1996, Performance changes and shareholder wealth creation associated
with mergers of publicly traded banking institutions, Journal of Money, Credit,
and Banking, 28, 284-310.
Pistor, K., M. Raiser and S. Gelfer, 2000, Law and finance in transition economies,
Economics of Transition 8, (2), 325-368.
Rapaczynski, A., 1996, The roles of the state and the market in establishing property
rights, Journal of Economic Perspectives, 10 Spring), 87-103.
Rhoades, S.A., 1993, Efficiency effects of horizontal (in-market) mergers, Journal of
Banking and Finance 17, 411-422.
Rostowski, J., 1995, The Banking System, Credit and the Real Sector in Transition
Economies, in J. Rostowski (editor), Banking Reform in Central Europe and the
Former Soviet Union, New York, (Oxford Economic Press, New York), 16-41.
Scholes, M., Williams J, 1977, Estimating betas from nonsynchronos data, Journal of
Financial Economics. 5(3):309-327.
Scholtens, B., 2000, Financial regulation and financial system architecture in Central
Europe, Journal of Banking and Finance 24, 525-553.
Smith, G. and E. Malkin, 2001, "Citigroup's surprise Mexican marriage." Business Week
Online (May 18).
Spindt, P.A., and Trahan, V., 1993, The impact of mergers on bank operating
performance, Working paper, Tulane University
Uhlenbruck, K., and De Castro, J.O., 2000, Foreign acquisitions in Central and Eastern
Europe: Outcomes of privatization in transitional economies, Academy of
Management Journal, 43(3), 381-402.
Unal, H., and Navarro, M., 1999, The technical process of bank privatization in Mexico,
Journal of Financial Services Research, 16(1) 61-83.
32
Verbrugge, J. A., Megginson, W.L., Owens, W.L, 1999, “State ownership and the
financial performance of privatized banks: An empirical analysis,” Paper
presented at the World Bank/Federal Reserve Bank of Dallas Conference on
Banking Privatization (March).
Wachtel, P., 1999, “Summary and introduction to special issue: Entrepreneurship in the
transition economies of Central and Eastern Europe,” Journal of Business
Venturing, 14, 5/6 (September/November), 17-25.
33
Table 1
Distribution of Financial Services Privatization Acquisition Announcements
Panel A. Distribution of announcements over time
Transactions
Announcement Year
1980-1990
6
Percent of Total
6.19%
1991
3
3.09%
1992
7
7.22%
1993
3
3.09%
1994
5
5.15%
1995
10
9.70%
1996
11
11.34%
1997
1998
15
15.46%
11
11.34%
1999
5
5.15%
2000
7
7.22%
2001
2002
7
7.22%
7
7.22%
Total
97
100
Panel B. Distribution of bidders
Number of
Announcements
55
Percent of
Announcements
56.77
42
43.23
97
100
47
48.45
Foreign
50
51.55
Total
97
100
Percent of Acquisition
Full (100%)
Partial (10 to 99%)
Majority
31
66
31.95
68.04
59
60.82
(Minority < 50%)
38
39.18
Total
97
100
Industry Sector
Depository institutions
Non-Banks
Total
Country
United States
34
Panel C. Distribution of privatizing financial services firms by region
Number of
Percent of
Industry Sector
Announcements
Announcements
Australia New Zealand
2
2.06
Central Asia
3
3.09
China
7
Europe
21
21.65
27
Latin and South America
27.84
Transition Economies
Southeast Asia
Total
25
25.77
12
12.37
97
100%
35
Table 2
Sample and Control Descriptive Statistics
Sample Firms
Mean
Size
Assets ($M)
Market Value ($M)
Profitability
ROA (%)
ROE (%)
NPM (%)
Performance – Bank Bidders
SGAX/ Total Assets (%)
Loans/Total Assets (%)
Nonperforming Assets/ Total
Assets (%)
Provision for Loan
Losses/Assets (%)
Net Charge Offs/Net Income
(%)
Control Firms
Mean
T-statistic
112,297
2.56**
33,077
15,030
3.56***
1.16
15.55
9.11
0.95
11.71
8.71
0.57
3.43***
0.53
3.02
42.70
2.30
47.44
2.09
0.92
-2.11**
1.05
0.47
1.91*
-298.88
2.39**
119,802
17.24
0.65
-34.93
This table shows mean descriptive statistics for both sample (bidding) firms and control firms for the year
prior to the announcement of the acquisition. Return on Assets (Equity) is the ratio of net income to total
assets (shareholders equity). Net profit margin is the ratio of net income to sales. SGAX/Total Assets is the
ratio of commercial expenses of operation incurred in the regular course of business pertaining to the
securing of operating income not allocated to costs of goods sold to total assets. Loans/total assets is the
ration of loans to total assets. Provision for loan losses/Assets is the “expense charged to earnings which
increases the allowance for possible losses on assets owned due to the decline in value of collaterized assets
or foreclosed assets and includes provision for loan losses, provision for other real estate owned losses,
specific and general provision for other real estate owned losses, and provision for loss on assets held for
sale (S&P Research Insights). Net Charge Offs/Net Income is the ratio of the “reported amount of asset
write-downs minus recoveries of previous write downs to net income. If losses exceed recoveries, this
value is shown as a negative amount”. Nonperforming assets/Total Assets represents the ratio of
nonperforming assets, defined as the reported amount of assets that are considered nonperforming,
including loans and leases carried on a non-accrual basis, loans which are 90 days past due both accruing
and non-accruing, renegotiated loans, real estate acquired through foreclosure, repossessed movable
property but excluding past due loans not yet placed on non-accrual status (unless specifically included in
nonperforming assets by the bank) divided by total assets. T-statistics are reported for the difference in
sample and control means for each variable.
*, ** and *** significant at the 10%, 5%, and 1% levels, respectively, using a two-tailed test.
36
Table 3
Cumulative abnormal returns
CAR Event Windows
No.
(-1, +1)
Panel A. Total Sample
All Bidders
97
Panel B. by Participant Type
US
47
US Nonbank
26
US Bank
21
Foreign
50
Nonbank
16
Bank
34
Bank Bidder
55
Non-Bank Bidder
42
(-1, 0)
+/-
0.61
0.27
2.20**
0.48
0.24
2.28**
58/39
2.35**
0.71
0.38
1.88*
0.86
0.64
1.98**
0.49
0.16
0.94
0.53
0.31
2.28**
0.79
0.23
2.21**
0.40
0.35
0.73
0.43
0.25
1.15
0.83
0.42
2.02**
0.33
0.15
1.11
0.77
0.49
1.57
-0.24
-0.69
-0.03
0.63
0.30
2.11**
0.97
0.89
1.76*
0.46
0.24
1.35
0.20
0.09
1.05
0.85
0.54
2.27**
25/22
0.69
14/12
0.61
11/10
0.48
33/17
2.52**
11/5
1.67*
22/12
1.93*
33/22
1.82*
25/17
1.50
This table presents the mean (first line) and median (second line) announcement period cumulative
abnormal returns (CARs) for the (-1, +1), (-1, 0) and (0) event windows for the 97 privatization
announcements. Abnormal returns are calculated using the market model estimated from 110 to 11 days
prior to the event announcements. CARs represent the cumulative market model-adjusted abnormal returns
over the relevant event window. The CRSP equally-weighted market index is used. The Z statistics (given
in parentheses) are based on the standardized cross-sectional method. The number of positive and negative
CARs for the (-1, +1) window (+/-) are reported in the last column, with the test statistic for the
nonparametric generalized sign test reported in parentheses under +/-.
*, ** and *** significant at the 10%, 5%, and 1% levels, respectively, using a two-tailed test.
37
R2
F
N
Constant
Majority
US = 1
Model 1
10.1
3.51***
89
0.052
(2.79)***
-0.005
(-1.05)
-0.012
(-2.48)**
Model 2
9.3
2.82**
89
0.052
(2.77)***
-0.005
(-1.08)
-0.011
(-2.45)**
-0.013
(-2.71)***
-0.003
(-2.23)**
-0.013
(-2.74)**
-0.003
(-2.26)**
Transition = 1
Bank/Nonbank
Ln (Market
Value)
Rule of Law
ROE
0.001
(0.46)
Table 4
Cross-Sectional Regressions
Model 3
Model 4
11.8
12.2
2.99**
2.38**
89
89
0.060
0.058
(3.16)***
(2.67)**
-0.008
-0.005
(-1.75)*
(-1.09)
-0.012
-0.001
(-2.59)**
(-2.42)**
-0.009
(-1.86)*
-0.015
-0.014
(-3.14)***
(-2.77)***
-0.004
-0.004
(-2.40)**
(-2.31)**
-0.001
(-0.52)
0.001
0.001
(0.64)
(0.50)
PCIit/PCIUSt
GDPit/GDPUSt
D(Total Risk)
Model 5
17.8
3.28***
89
0.076
(4.15)***
-0.009
(-1.95)*
-0.011
(-2.56)**
-0.010
(-1.96)*
-0.019
(-4.04)***
-0.005
(-3.09)***
-0.002
(-0.89)
0.001
(0.24)
-0.001
(-0.470)
0.032
(0.77)
Model 6
13.9
3.05**
89
0.058
(3.01)***
-0.007
(-1.34)
-0.012
(-2.64)**
-0.008
(-1.60)
-0.014
(-2.95)***
-0.004
(-2.52)**
Model 7
12.4
2.80**
89
0.061
(3.21)***
-0.009
(1.80)*
-0.013
(-2.78)***
-0.010
(-1.99)**
-0.016
(-3.22)***
-0.004
(-2.44)**
0.001
(0.75)
0.001
(0.71)
0.003
(2.03)**
D (Beta)
-0.004
(-1.23)
38
This table provides the results of cross sectional regressions for the full sample of 97 financial services privatization acquisitions. The dependent variable is the
(-1,0) cumulative normal return, where 0 is the announcement date and -1 is the day prior to the announcement date. Return on Equity is defined as the ratio of
net income to total equity for the acquiring firm for the year prior to the acquisition. Ln(Market Value) is the natural log of market value in the year prior to the
acquisition. Early is a dummy variable equal to 1 if the acquisition took place prior to 1996 and 0 otherwise. Bank/Nonbank is a dummy variable equal to 1 if the
bidder is a commercial bank (sic codes 6021, 6022, or 6029) and 0 otherwise. Transition is a dummy variable equal to 1 if the privatizing target is in an Eastern
Europe transition economy (Hungary, Poland, the Czech Republic, Russia, Moldova, Kazakhstan, Russia, Ukraine). Majority is a dummy variable equal to 1 if
the acquisition is over 50% of the shares of the privatizing firm. US is a dummy variable equal to 1 if the bidder is a US firm. GDPit/GDPUSt is nominal Gross
Domestic Product in the home country of the acquired privatized bank divided by US GDP for the same year . PCIit/PCIUst is the same measure for percapita
GDP. Rule of Law is a commercial political risk management index measuring on a one (lowest) to six scale the extent to which contracts and private property
rights are governed and protected by the rule of law. D(Total Risk) and D(Beta) are the changes in total risk and systemic risk, respectively, for the acquiring
firm from before to after the merger, as estimated through the regression equations, and as described in Table 5.
*, ** and *** denote significance at the 10%, 5%, and 1% levels, respectively, using a two-tailed test.
39
Table 5
Changes in Risk
Sample
Total
Risk
Beta
0.237
-0.059
(2.91)***
(-0.75)
0.287
-0.075
(2.39)**
(-1.69)*
0.171
-0.016
(1.69)*
(-0.13)
0.182
-0.073
(1.81)*
(-0.65)
0.306
-0.119
(1.70)*
(-2.19)**
0.087
-0.033
(0.80)
(-0.17)
0.301
-0.022
(2.33)**
(-0.36)
0.273
-0.041
(1.68)*
(-1.48)
0.347
0.008
(1.62)
(0.10)
Number
Panel A. Total Sample
All Bidders
94
Panel B. by Participant Type
US Bidder
47
Foreign Bidder
47
Bank Bidder
53
US Bank Bidder
25
Foreign Bank Bidder
28
Non-Bank Bidder
41
US Non Bank
22
Foreign Non Bank
19
This table reports the changes in total risk and systematic risk or beta around the announcement of the
acquisition announcement. For each bidder, we estimate the pre-announcement variable, VARPRE [postannouncement variable, VARPOST ] over the period from t-110 to t-11 [t+11 to t+110], where t=0 is the
announcement date. The difference is calculated as VARPOST -VARPRE . A positive [negative] change in
the bank’s risk measure represents an increase [a decrease] in risk over the measurement period. Tstatistics are reported in parentheses.
* and **significant at the 10% and 5% levels, respectively, using a two-tailed test.
40
Table 6
Long-Horizon Average Holding-Period Returnsa
Maximum
Holding Period Return Event
Windows, %
Sample Type
No.
(+1, +6)
Panel A. Total Sample
92
-1.88
(-0.30)
Overall Sample
(+1, +12)
(+1, +18)
-6.86
(`-1.42)
-10.06
(-1.79)*
-11.56
(-1.44)
-2.65
(-0.48)
-14.31
(-2.06)**
-23.09
(-1.88)*
-7.18
(-0.96)
5.15
(1.04)
5.08
(0.78)
5.22
(0.67)
-13.98
(-1.69)*
-6.61
(-0.87)
-14.29
(-1.71)*
-24.12
(-1.96)*
-6.61
(-0.62)
-3.45
(-0.50)
0.11
(0.01)
-7.45
(-0.75)
Panel B. By Bidder Characteristics
US Bidder
45
Foreign Bidder
47
Bank Bidder
53
US
Foreign
25
28
Nonbank Bidder
US
Foreign
39
20
19
-13.03
(-1.54)
8.77
(1.80)*
-4.11
(-0.67)
-19.84
(-1.56)
6.42
(0.97)
1.08
(0.12)
-8.96
(-0.81)
12.36
(1.70)*
a
This table reports the long-horizon average holding-period abnormal returns (AHAR) for our sample
banks. We compute long-horizon holding period raw returns for the sample firms (HPRFi) and for the
matched firms (HPRCi). AHARi = HPRFi - HPRCi. HPRFi, HPRCi and AHARi are calculated starting
with the month after the announcement date for 6, 12, and 18 month holding periods. Banks in the sample
with multiple transactions that confound the long-horizon analysis are omitted from the analysis. Further,
due to size considerations, we are unable to match certain banks in the sample. Thus, the sample sizes in
these tests are smaller than those in the near-term event study. The t statistics are given in parentheses.
***, **, *Significant at the 1%, 5%, and 10% levels, respectively.
41
Table 7
Percent Change in Operating Performance After Acquisition
Profitability
ROA (%)
ROE (%)
NPM (%)
Performance – Bank Bidders
SGAX/ Total Assets (%)
Loans/Total Assets (%)
Nonperforming Assets/ Total
Assets (%)
Net Charge Offs/Net Income
(%)
Change for
Sample Firms
Change for
Control Firms
T
(Sample –
Control)
-0.121
-0.094
-0.064
0.059
0.101
0.127
1.63
1.66*
1.80*
0.491
0.012
0.269
-0.058
-0.446
-1.54
-1.59
-3.74***
-0.259
-1.75*
0.134
0.446
Variables shown are computed as the pre-announcement variable, VARPRE [post-announcement variable,
VARPOST ] over the period from t +0 to t-2years [t+0 to t+2 years], where t=0 is the announcement date.
The change is calculated as (VARPOST -VARPRE )/VARPRE for the firms participating in these privatization
transactions. Return on Assets (Equity) is the ratio of net income to total assets (shareholders equity). Net
profit margin is the ratio of net income to sales. SGAX/Total Assets is the ratio of commercial expenses of
operation incurred in the regular course of business pertaining to the securing of operating income not
allocated to costs of goods sold to total assets. Loans/total assets is the ration of loans to total assets.
Provision for loan losses/Assets is the “expense charged to earnings which increases the allowance for
possible losses on assets owned due to the decline in value of collaterized assets or foreclosed assets and
includes provision for loan losses, provision for other real estate owned losses, specific and general
provision for other real estate owned losses, and provision for loss on assets held for sale (S&P Research
Insights). Net Charge Offs/Net Income is the ratio of the “reported amount of asset write-downs minus
recoveries of previous write downs to net income. If losses exceed recoveries, this value is shown as a
negative amount”. T statistics are reported to represent the change in variable for both sample and control
firms, and for the difference in sample and control means for each variable.
*, ** and *** significant at the 10%, 5%, and 1% levels, respectively, using a two-tailed test.
42
Table 8
Multivariate Analysis of 12 Month Long-Horizon Returns
Rsq
F
N
Constant
US = 1
Rule of Law
Model 1
Model 2
Model 3
6.00
2.91**
90
-0.156
(-0.94)
-0.105
(-1.09)
0.061
(1.89)*
16.50
3.93***
84
-0.047
(-0.23)
-0.001
(-0.01)
0.084
(2.46)**
0.044
(0.42)
-0.229
(-2.15)**
14.20
2.2888
84
-0.082
(-2.21)**
-0.030
(-0.36)
0.090
(2.33)**
0.016
(0.19)
-0.058
(-0.73)
-1.336
(-1.71)*
-0.001
(-1.32)
0.083
(3.09)***
-0.017
(-2.89)***
Majority
Bank/Nonbank
-0.173
(-1.73)*
GDPit/GDPUst
PCIit/PCIUSt
Ln(Market Value)
ROE
-0.225
(-3.04)***
This table presents the results of multivariate regressions on 12 month control adjusted buy and hold
returns for the sample of 52 acquisitions for which data was available. Asset Growth and Change in ROE
are computed as the pre-announcement variable, VARPRE [post-announcement variable, VARPOST ] over the
period from [t+0 to t+2 years], where t=0 is the announcement date. The change is calculated as (VARPOST
-VARPRE )/VARPRE for the firms participating in these privatization transactions. US = 1 if the bidder is a
US company, 0 otherwise. Majority is if the percent of the acquisition was 50% or greater. Bank/Nonbank
=1 if the acquirer is a bank, 0 otherwise. Ln(Market Value) is the natural log of market value in the year
prior to the acquisition. Bank/Nonbank is a dummy variable equal to 1 if the bidder is a commercial bank
(sic codes 6021, 6022, or 6029) and 0 otherwise. Majority is a dummy variable equal to 1 if the acquisition
is over 50% of the shares of the privatizing firm. US is a dummy variable equal to 1 if the bidder is a US
firm. GDPit/GDPUSt is nominal Gross Domestic Product in the home country of the acquired privatized
bank divided by US GDP for the same year. PCIit/PCIUst is the same measure for percapita GDP. Rule of
law is a commercial political risk management index measuring on a one (lowest) to six scale measuring
the extent to which contracts and private property rights are governed and protected by the rule of law.
*, ** and *** significant at the 10%, 5%, and 1% levels, respectively, using a two-tailed test.
43