Article

The Choice of Exit: Journal of Emerging Market Finance

17(1S) 1–26

Influence of Private © 2018 Institute for Financial

Management and Research
SAGE Publications
Equity Investors and sagepub.in/home.nav
DOI: 10.1177/0972652717751534
Buyout Entry http://journals.sagepub.com/home/emf

Rohan Chinchwadkar1
Rama Seth2

Abstract
The choice of exit method is an inevitable decision faced by entrepre-
neurs and private equity (PE) investors. The existing literature addresses
four categories of factors which influence this choice of exit method
between initial public offering (IPO) and acquisition: industry-related
factors, market-timing variables, deal-specific factors and demand-for-
funds factors. We extend the literature by introducing a new category
of factors, ‘PE investor characteristics’, and test if this category has a
significant effect on the choice of exit method. We also test if the type
of entry has an influence on the exit method. We find that PE investor
characteristics play an important role in the choice of exit method. The
existence of a large syndicate of PE investors in the same firm increases
the probability of an IPO exit, but the presence of a foreign PE inves-
tor reduces this probability. Moreover, unlike in developed markets,
the cost of debt does not affect the choice of exit method in India. We
further consider specific exit methods such as strategic sale, financial
sale and buyback and find consistent results. We find that in buyout trans-
actions, the probability of an IPO exit is less than that of a strategic sale.

1
Assistant Professor, Indian Institute of Management Trichy, Tiruchirapalli, Tamil Nadu,
India.
2
Professor, Indian Institute of Management Calcutta, Joka, Kolkata, West Bengal, India.

Corresponding author:
Rohan Chinchwadkar, Assistant Professor, Indian Institute of Management Trichy,
Tiruchirapalli 620015, Tamil Nadu, India.
E-mails: rohan.chinchwadkar@gmail.com; rohan@iimtrichy.ac.in
2 Journal of Emerging Market Finance 17(1S)

Finally, we present a unique finding that the probability of a buyback as

opposed to an IPO is higher if a firm is in the real estate sector.

Keywords
Initial public offerings, acquisitions, private equity exits, buyout, informa-
tion asymmetry

1. Introduction
Private equity (PE) investors typically provide capital to private compa-
nies, usually for expansion, new product development or restructuring of
the company’s operations, management or ownership. As the firm grows,
PE investors sell their stakes in the company either to return the capital
to the limited partners or to find new investee companies. Thus, in the
investment cycle, a PE investor realises return on an investment only
after a successful exit. There are four major options of exit for PE inves-
tors: initial public offering (IPO), financial sale, strategic sale and buy-
back. The major difference between IPO and other methods is that an
IPO involves a large number of dispersed investors whereas the other
three methods involve a single or very few investors. Thus, the other
three categories are typically considered under a single category of
‘acquisitions’. The objective of this article is to study the factors which
affect the choice of exit method between IPO and acquisitions. We also
analyse the factors which affect specific types of acquisitions such as
financial sales, strategic sales and buybacks.
The rest of this article is organised as follows. We present a detailed
review of the relevant literature in Section 2. Section 3 discusses the
factors which affect the choice of exit between IPO and acquisition as
well as between specific types of acquisitions such as strategic sale,
financial sale and buyback. We develop hypotheses about how PE inves-
tor characteristics affect the choice of exit method. Section 4 analyses the
data set and presents descriptive statistics. Multivariate empirical models
and analysis of results are presented in Section 5 which also presents an
analysis of the marginal effects of independent variables on all the exit
methods. Section 6 concludes this article.

2. Literature Review
In this section, we review the relevant literature from three areas: IPOs,
acquisitions and PE exits.
Chinchwadkar and Seth 3

2.1. IPO: The Decision to Go Public

The theory of the decision to go public has been studied extensively in
literature. Ritter and Welch (2002) present an excellent review of litera-
ture related to IPOs. They analyse three themes related to IPOs: why
firms go public, why there is existence of IPO underpricing and why IPOs
exhibit underperformance in the long run. In this article, we focus on the
first question. The broad answer to this question is: ‘To raise equity capi-
tal for the firm and to be part of the public market in which the founders
and other existing shareholders can convert some of their wealth into cash
at a future date’. There are largely two schools of thought in literature that
answer why firms choose to go public: life cycle theories (Black & Gilson,
1998; Chemmanur & Fulghieri, 1999; Maksimovic & Pichler, 2001;
Schultz & Zaman, 2001; Zingales, 1995) and market timing theories
(Benninga, Helmantel, & Sarig, 2005; Korajczyk, Lucas, & McDonald,
1992; Subrahmanyam & Titman, 1999).

Life cycle theories: The first formal theory of the going public decision
was presented by Zingales (1995), who observed that it is easier for an
acquirer to spot a potential takeover target when it is public. Also, firms
realise that acquirers can better pressure targets on pricing discounts
when they are private rather than when they are public. Thus, by going
public, entrepreneurs would be able to demand a higher price for their
company than what they would get from a direct sale. In contrast to the
this exit-oriented view, Black and Gilson (1998) find that an IPO is the
most common method for entrepreneurs to gain control of the firm from
exiting venture capitalists. Chemmanur and Fulghieri (1999) discuss the
advantages and disadvantages of an IPO from the point of view that an
IPO creates dispersed ownership. They study the fact that pre-IPO inves-
tors usually hold undiversified portfolios and, therefore, are not willing
to pay as high a price as public market investors who hold diversified
portfolios. However, there are fixed costs of going public and informa-
tion cannot be revealed to small, dispersed investors without any cost.
Thus, early in the lifecycle of a firm, it will be privately held, and when
it becomes sufficiently large, it becomes optimal to go public in the later
stages of its life cycle.

Market timing theories: This set of theories is based on the assumption

that firms try to ‘time’ the market to maximise the valuation which they
will receive in the public market. Korajczyk et al. (1992) develop a model
of the effect of asymmetric information to show that firms postpone the
4 Journal of Emerging Market Finance 17(1S)

timing of their IPOs if they are currently being undervalued and wait for
bull markets to issue equity. Subrahmanyam and Titman (1999) suggest
that the value of public firms increases with the size of public market and
thus when the stock market in a particular economy reaches a critical
mass, it causes the IPO market to ‘snowball’, that is, new firms list on the
stock market, making the market more liquid and efficient, which in turn
induces more firms to go public. Benninga et al. (2005) endogenise the
timing of the decision to go public and explain some observed phenomena
such as clustering of IPOs and buyouts, and the industry concentration
of IPO waves.

2.2. Choice between IPO and Acquisition

The literature which focuses only on IPOs addresses only two outcomes:
staying private and going public via IPO. However, a firm often faces a
different kind of choice: going public via IPO or selling the firm to an
acquirer. The motivation of the acquirer could be rooted in strategic or
financial reasons. There have been very few studies which analyse
the choice between IPO and acquisition as alternative exit options
(Bayar & Chemmanur, 2012; Brau, Francis, & Kohers, 2003; Poulsen &
Stegemoller, 2008). Brau et al. (2003) define four categories of factors
which influence a firm’s choice between IPO and acquisition: industry-
related factors, market-timing variables, deal-specific factors and demand-
for-funds factors. Poulsen and Stegemoller (2008) study the effect of
firm characteristics such as growth and capital constraints on the choice
of exit method.

2.3. The IPO versus Acquisition Decision in PE-backed Firms

The literature which addresses the choice of exit between IPO and acqui-
sition has dealt with generic firms. However, the choice of exit becomes
more complicated in the presence of PE investors who invest in small,
privately held firms to help them grow and then realise returns through
the ‘exit’ transaction (Cumming & MacIntosh, 2003; Farag, Hommel,
Witt, & Wright, 2004; Fleming, 2004; Kutsuna, Cowling, & Westhead,
2000; McKaskill, Weaver, & Dickson, 2004; Neus & Walz, 2005;
Parhankangas, Landström, & Smith, 2005).
It has been found that returns vary substantially across different exit
outcomes such as IPO, financial sale, strategic sale and buyback (see,
Chinchwadkar and Seth 5

e.g., Cochrane, 2005; Nikoskelainen & Wright, 2007). Apart from finan-
cial returns, certain exit outcomes produce reputation benefits. Gompers
(1996) finds that younger PE firms prefer to take their portfolio companies
public through an IPO because the reputation benefits of the IPOs help
them during the next round of fund raising.
A number of factors affect the exit outcome of PE firms. Cumming
and Johan (2008) study the theory and evidence connecting information
asymmetry and agency costs to exit outcomes in PE-backed firms. They
mention that where the PE investors are better able to mitigate information
asymmetry and agency costs faced by new owners, they are more likely
to have successful exit outcomes.
IPO is not only the most difficult exit outcome to achieve since
information asymmetry is the highest but also offers the highest returns.
Information asymmetry is lower for a financial sale than an IPO since the
buyers are sophisticated financial investors who can conduct due diligence.
It is even lower in the case of a strategic sale since the buyers in stra-
tegic sales are typically large players from the same industry who have a
higher level of expertise and insight about the industry. Buybacks have
no information asymmetry problem since the buyers are the promoters
themselves but are also usually the least profitable since no new capital
enters the firm. Thus, we come to a hierarchy of exit outcomes according
to difficulty of information asymmetry resolution (Figure 1).
In this article, we contribute to the existing literature by testing the
effect of a new category of factors on the choice of exit method between
IPO and acquisition: PE-investor characteristics (variables such as size
of PE syndicate and type of PE investors). Specifically, we analyse how

Figure 1. Hierarchy of Exit Methods by Difficulty of Information Asymmetry

Resolution
Source: Povaly (2007), slightly modified.
6 Journal of Emerging Market Finance 17(1S)

the presence of multiple PE investors, foreign PE firms and co-investment

by foreign and domestic PE firms affects the choice of exit method. We
also test the effect of these parameters on detailed exit outcomes: IPO,
strategic sale, financial sale and buyback. Further, we test the influence of
‘type of entry’ on the exit method. Specifically, we test if buyout transac-
tions have a significant effect on the exit method.

3. Hypotheses Development
A number of factors affect the exit method of a PE-backed firm. It depends
on the ability of the new owners to resolve information asymmetry
problems and to get an accurate estimate of the value of the firm. While
examining the choice of exit method between IPO and acquisition,
Brau et al. (2003) consider four categories of factors which affect the
attractiveness of an IPO as compared to an acquisition: industry-related
factors, market-timing variables, deal-specific factors and demand-for-
funds factors. Cumming and Johan (2008) discuss the effect of venture
capital characteristics (such as ownership structure, province and stage,
among others), transactional characteristics and firm characteristics on
various exit outcomes: IPO, financial sale, strategic sale and buyback.
Each category of variables improves or worsens information asymmetry
problems.

3.1. PE Investor Characteristics

We now introduce a new category of variables (PE investor characteris-
tics) and hypothesise its effect on the choice of exit method.
3.1.1. Size of PE Syndicate: Number of PE Investors
The presence of multiple PE investors in the same firm is a common
phenomenon. For example, three PE firms, Intel Capital, Norwest Venture
Partners and Gabriel Partners, invested in Persistent Systems, a Pune-
based IT firm, and all of them part-exited during the IPO. The presence
of multiple PE investors in the same firm usually signifies that due dili-
gence of the firm has been done multiple times and the firm has been
found to be attractive. Also, more PE investors can mean that more value
has been added to the firm due to higher involvement of investors. Thus,
more PE investors in a firm may signal a better quality of a firm.
When it is time to exit, multiple investors present certain problems.
First, the liquidity requirements of all PE investors may be different from
Chinchwadkar and Seth 7

each other. Second, young PE firms might prefer IPO over acquisition due
to reputation benefits created by an IPO (Gompers, 1996). Third, every PE
investor may have a different view about synergy and expected valuation
of the firm. Game theory research suggests that equilibrium in multi-party
negotiation is much more complex and difficult to achieve than in a two-
party negotiation. These considerations make it difficult for the firm to
negotiate with the acquirers in the presence of multiple PE investors.
However, when it comes to an IPO, the process is standard. The price band
of the IPO is decided in collaboration with the investment banker, and
feedback is received from the investors during the book-building process.
Also, after the launch of IPO, all PE investors possess shares of a liquid
and listed entity. At the same time, a larger PE syndicate improves the
capability of the exiting investors to participate in larger roadshows for
IPOs and resolve information asymmetry with a larger group of buyers.
Thus, it is easier for large syndicates to execute IPOs. This leads us to
the following hypothesis:

H1: The probability of an IPO exit increases with increase in number

of PE investors exiting from the firm.

3.1.2. Foreign Investors: Presence of Foreign PE Investors

PE firms from developed markets have been increasingly investing in
firms in developing markets like India, especially since the beginning of
the twenty-first century. We can see multiple examples of such invest-
ments: Intel Capital in 123Greetings.com, JP Morgan in Binani Cement,
Temasek in Infinite Computer Solutions and many more. Cumming and
Johan (2006) mention that if the investee firm and PE investors are from
the same province, it enhances the ability of the PE investor to conduct
due diligence and reduces adverse selection problems. Thus, foreign PE
investors are not able to resolve information asymmetry problems as
effectively as domestic PE investors. Since IPOs need a higher capability
of PE investors to resolve information asymmetry, it follows that going
for an IPO might be more difficult for foreign PE firms than domestic
ones. This leads us to the following hypothesis:

H2a: Foreign-PE-investor-backed firms are less likely to prefer an

IPO over an acquisition.

However, many a times, foreign and domestic PE firms form syndicates

and invest together in certain target firms. The investments by New Vernon
Capital and Bessemer Venture Partners in Motilal Oswal, ChrysCapital
8 Journal of Emerging Market Finance 17(1S)

and J P Morgan in Titagarh Wagons, Sequoia Capital India and Actis in

Paras Pharmaceuticals are examples of such co-investments. If there is
co-investment by foreign and domestic PE investors in the same firm,
the ability of foreign PE investors to resolve information asymmetry
problems increases due to two reasons: presence of multiple investors in
the syndicate and presence of domestic PE investors in the syndicate.
Thus, information asymmetry problems in an IPO reduce due to a co-
investment structure.

H2b: Firms which have co-investment by both foreign and domestic

PE investors are more like to prefer an IPO over an acquisition.

To test the aforementioned hypotheses, we divide exits into three types,

depending on the country of origin of the PE firm: (a) foreign (invest-
ment made by foreign PE firm), (b) domestic (investment made by
domestic PE firm) and (c) co-investment (both foreign and domestic PE
firms have invested in the firm).

3.2. Industry Characteristics

It is known that certain industries have more serious information asym-
metry problems than others. Also, synergies are more important in some
industries and not very significant in others.
The financial services industry is highly regulated and is characterised
by high synergies. Since it is a relationship-oriented service industry, there
is an existence of soft assets. This creates a high level of information asym-
metry. The presence of high synergies and high information asymmetry
leads us to the conclusion that financial services firms are more likely
to go for acquisitions than IPOs. Apart from this, Berger, Demsetz and
Strahan (1999) show that financial services firms are able to achieve higher
profit efficiency, diversification of risks and higher efficiency of payments
systems through acquisitions. This leads us to the following hypothesis:

H3: Financial services firms are more likely to prefer acquisitions

over IPOs.

Industries such as life-sciences and pharmaceuticals significantly depend

on high-tech research and development. These industries are thus char-
acterised by high information asymmetries. Yosha (1995) shows that
entrepreneurial firms in high-tech industries prefer bilateral methods of
Chinchwadkar and Seth 9

financing to multilateral options in order to avoid leakage of sensitive

information to competitors. Also, we know that high-tech firms are able
to attract relatively higher acquisition premiums than firms in other
industries (N. Kohers & T. Kohers, 2000). This leads us to the following:

H4: High-tech firms are more likely to prefer acquisitions over IPOs.

PE transactions in the real estate sector are very different from those in
other sectors. This is mainly because most real estate investments are in
‘special purpose vehicles’ or SPVs rather than in typical companies.
An example of such a SPV is Delhi International Airport Limited (DIAL)
which was created by the contractors who helped build and maintain the
Delhi International Airport. Annamalai and Doshi (2012) study PE
investments in the real estate sector in India and find that almost 70 per
cent of the investments were at the project or SPV-level. Since such
SPVs typically do not fit the criteria of a ‘going concern’, IPOs are very
rare in the real estate sector. Also, in typical residential real estate pro-
jects, promoters are cash-rich at the end of the projects, due to inflows
from bookings and purchases. Annamalai and Doshi (2012) also mention
that the major reason for SPV-level investments is the low degree of
transparency and high level of information asymmetry that characterises
the sector. Thus, we would expect buybacks to be a much easier and
preferred exit choice in real estate investments. This brings us to a unique
hypothesis about the real estate sector:

H5: Real estate firms are more likely to prefer buybacks over IPOs.

Industry Valuation and Concentration

There is mixed evidence about the effect of industry-wide valuations on
the decision of firms to conduct IPOs or acquisitions. Golbe and White
(1993) analyse merger waves and show that the valuation of a target may
be based on the average valuation of the industry as captured by industry
average market-to-book (M/B) ratio. Thus, a higher industry M/B would
help targets in that industry to attract a higher acquisition premium. On
the other hand, Pagano and Roell (1998) say that a higher industry M/B
creates a favourable environment for IPOs in that industry. Thus, we test
the following hypothesis to see which theory is supported by our data.

H6: Higher the industry market-to-book ratio, more is the probability

of acquisitions versus IPOs.
10 Journal of Emerging Market Finance 17(1S)

Industry structure is an important determinant of firm behaviour.

The challenges faced by a new entrant depend heavily on the structure of
the industry. Sharma and Kesner (1996) study the link between pre-entry
conditions of an industry and the post-entry performance of a new
entrant. Among other things, they find that the post-entry performance
and survival probability of a new entrant is negatively correlated with
the concentration of the industry. Thus, we can say that, in highly con-
centrated industries, firms would prefer to sell off to an existing large
incumbent instead of going public and competing with the large
incumbents.

H7: Higher the concentration of the industry, lower is the probability

of an IPO exit as compared to acquisition.

3.3. Deal-specific Characteristics

In this section, we focus on the effect of deal-specific characteristics on
the exit outcome.
3.3.1. Size of the Exit/Deal
It is known that IPOs have higher fixed costs due to compliance, invest-
ment bankers and other process-specific reasons. It is easier for a larger
firm to afford these fixed costs rather than a small firm. The higher fixed
costs component of an IPO discourages small firms from going public
(Holmstrom & Tirole, 1993; Ritter, 1987). Also, being a very large firm
narrows the set of potential acquirers. Thus, we would expect larger
firms to go public rather than be acquired.

H8: Larger the deal-size, more would be the probability of an IPO

exit as opposed to an acquisition.

3.3.2. Buyout Transactions

It has not been explored in the literature, how or whether the type of
entry in a PE transaction has an influence on the type of exit. KPMG
(2011) report titled Returns from Indian Private Equity analyses the
effect of entry type on the type of exit and notes that in a buyout, the sole
decision-maker is the PE firm. Adding a signalling dimension, it says, ‘It
is natural that buyouts will exit through strategic sales because the inves-
tors control the company. If the PE funds own a majority and exit through
IPO, the market will worry that there is something wrong with the com-
pany.’ This leads us to the following hypothesis:
Chinchwadkar and Seth 11

H9: In buyout transactions, the probability of an IPO exit is lower as

compared to acquisition.

3.3.3. Effect of the Global Financial Crisis

Existing market conditions severely affect a firm’s decision to go public,
especially in times of crisis. We test if, over and above the existing
market conditions (proxied by market return), the global financial crisis
had an effect on the choice of exit between IPO and acquisition in India.
We choose the year 2009 (variable is named ‘recession year’) because
our univariate tests show that the ratio of IPO exits to acquisition exits
was significantly different in this year. This is not a proxy for ‘recession’
in its usual definition of growth contraction, but a proxy for a year when
capital markets in India were severely affected due to spill-over effects
of recessions in developed markets during the global financial crisis.

H10: In the recession year (2009), probability of IPO exits will be

lower than acquisitions.

3.4. Market-related Factors

It is known that firms try to ‘time’ the markets by conducting IPOs in the
so-called hot markets when the existing market conditions are favourable
(Helwege & Liang, 2004; Lowry & Schwert, 2002; Pástor & Veronesi,
2005; Schultz, 2003). Thus, we examine the effect of current and lagged
month market return on the exit outcome (lagged-returns up to 12 months
before the deal were included in the regression and only the statistically
significant lags have been reported). Since higher market returns signify
a better market environment, we expect higher market returns to be
followed by more IPO exits than acquisitions.

3.5. Demand-for-funds Factor (Cost of Debt)

This category mainly analyses the effect of the cost of debt on the choice
of exit method. According to the pecking order theory, firms raise capital
in this preference order: internal accruals, external debt financing and
finally external equity financing. Thus, the higher cost of debt makes
IPOs more attractive as can be deduced from the pecking order theory
(Myers & Majluf, 1984). At the same time, higher cost of debt makes
it costlier to arrange funding for acquisitions (Golbe & White, 1993).
12 Journal of Emerging Market Finance 17(1S)

Thus, the higher cost of debt would result in higher IPO exits as com-
pared to acquisitions. In this article, we measure the cost of debt by the
3-month T-Bill rate (following Brau et al., 2003) because the cost of short-
term debt is especially relevant for bridge-loans provided for leveraged
acquisitions.

4. Sample Selection and Descriptive Statistics

In this section, we present the data set and sample used for the empirical
tests, along with univariate tests of difference.

4.1. Data
The major databases used for this study are Venture Intelligence and
Centre for Monitoring Indian Economy (CMIE) Prowess. We analyse a
sample of 447 PE exits in India from 2004 to 2010, out of which 97 are
IPOs and 350 are acquisitions. As mentioned earlier, we use this sample
period because it represents the first cycle in the Indian PE market
(Figure 2). Figure 3 shows the ratio of PE exits through acquisitions to
IPOs from 2004 to 2010. We see that this ratio hits a peak in 2009 when
the IPO market volume was low. Appendix A has a detailed explanation
of the variables and the sources.
Table 1 presents the deal sizes from 2004 to 2010 and analyses how
many of them were exits via IPOs and how many were acquisition exits.
We see that IPO exits were at their peak in the boom year of 2007 but
dipped significantly in the recession year of 2009. Table 2 analyses how
the exits have been distributed across industries. We see that the top three
industries by deal value during this period were IT and ITES, banking
and financial services, and telecom. The industries which experienced the
least exits according to the deal value were mining and minerals, educa-
tion and agri-business.

4.2. Descriptive Statistics

Table 3 presents the descriptive statistics of the independent variables.
We see that for the 447 exits being studied, the average number of
PE investors in each exit was about 1.5; around 50 per cent of the exits
involved at least one foreign PE firm and around 12 per cent of the exits
involved co-investment between foreign and domestic PE firms.
Chinchwadkar and Seth 13

Figure 2. PE Investments in India (USD Bn)

Source: Venture Intelligence.

Figure 3. Ratio of PE Exits via Acquisitions to IPOs from 2004 to 2010

Source: Venture Intelligence.

According to the number of exits, 8 per cent were in financial services

and 48 per cent exits involved firms from high-tech industries. The data
for deal size are available for only 336 deals and the average deal size
was around $78 million.
Table 4 represents the correlation matrix of the independent variables
of our empirical model. We see that the correlation between ‘number of
14 Journal of Emerging Market Finance 17(1S)

Table 1. Distribution of Deals from 2004 to 2010

Number of Deal Size

Year Exits (US$ million) IPOs Acquisitions
2004  32 1495.18  7  25
2005  52 4090.96 15  37
2006  41 2635.9 12  29
2007  79 3047.24 18  61
2008  44 4741.32 10  34
2009  66 2565.82  7  59
2010 133 7681.47 28 105
Source: Venture Intelligence.

Table 2. Distribution of Deal Size across Industries

Deal Size Deal Size

Industry (US$ million) Industry (US$ million)
Agri-business 85.7 Manufacturing 1389.36
Banking and Financial 4391.5 Media and 734.22
Services Entertainment
Education 45.14 Mining and Minerals 27.3
Energy 1816.1 Other Services 34
Engineering and 1551 Retail 205.2
Construction
FMCG 102.8 Shipping and Logistics 551.45
Food and Beverages 185 Telecom 3083.9
Gems and Jewellery 83.5 Textiles and Garments 121.42
Healthcare and Life 2675.9 Travel and Transport 603.7
Sciences
Hotels and Resorts 148.02 Real Estate 2743.2
IT and ITES 5679.48
Source: Venture Intelligence.

PE investors’ and ‘coinvestment dummy’ is very high. Thus, we will use

separate empirical models to test our hypotheses about these variables to
avoid the problem of multi-collinearity.

4.3. Univariate Tests

We first run univariate tests of difference between means and medians.
We calculate the means and medians of the IPO sample and acquisition
sample for all the five categories of independent variables mentioned.
Table 5 shows the results of univariate difference tests (t-test and
Chinchwadkar and Seth 15

Table 3. Descriptive Statistics

Independent Variables N Mean Std. Dev.

Number of PE Investors 447 1.42953 0.887797
Foreign PE Dummy 447 0.496644 0.500549
Co-investment Dummy 447 0.120805 0.326266
Financial Services 447 0.082774 0.275849
High Tech 447 0.487696 0.500409
Industry Market-to-book 446 3.581749 2.002894
Industry Herfindahl Index 446 0.080221 0.077289
Deal size (US$ million) 336 78.14848 185.5878
Market Return (current month) 447 2.355454 6.917677
Market Return (previous month) 447 2.862287 7.204161
3-month T-Bill Rate 447 5.52519 1.490336
Source: Venture Intelligence.

Wilcoxon test). An initial inspection of the table shows that the mean
and median of the IPO and M&A sample are significantly different for
all variables of the ‘PE investor characteristics’ category. The mean
number of PE investors for the IPO sample is 1.9897 which is greater
than that of the acquisition sample which is 1.2675, at 1 per cent level of
significance. Similarly, the involvement of foreign PE investors in IPO
exits is significantly higher than that in the acquisition sample. When we
analyse the results of univariate tests further, we see that 26 per cent of
IPOs involved a co-investment between foreign and domestic PE firms
which is significantly higher than the 7 per cent in acquisitions. Thus, on
primary inspection, we observe that on average IPO exits involve more
number of PE investors, more foreign PE firms and more co-investments
as compared to acquisitions.

5. Empirical Model and Regression Results

After analysing univariate tests, we move on to multivariate tests of our
hypotheses.

5.1. Logistic Regression Results

Since the dependent variable is binary (IPO or acquisition), we use logis-
tic regressions and estimate Model (1):
Table 4. Correlation Matrix of Independent Variables

Real Market Market

Number PE Foreign Co-invest Fin Ser High-tech Estate Deal Size Buyout (t) (t – 1) T-Bill
Number PE 1
Foreign Dummy 0.3301 1
Co-invest Dummy 0.722 0.3708 1
Fin Services 0.1213 0.1151 0.1223 1
High-tech 0.05 0.0707 0.051 – 0.298 1
Real Estate –0.0482 0.0215 –0.0103 –0.0774 –0.2185 1
Deal Size 0.2922 0.2052 0.2114 0.1031 0.0909 0.0951 1
Buyout Dummy 0.0744 –0.0456 0.1178 –0.0751 –0.0482 –0.0551 –0.0437 1
Market (t) –0.0644 –0.0349 –0.0709 –0.0373 0.0297 –0.0648 –0.0292 –0.0111 1
Market (t – 1) –0.0765 –0.041 –0.1171 0.0023 0.0099 –0.0922 –0.1068 0.0237 0.0404 1
3M T-Bill Rate 0.0262 –0.105 0.0316 –0.1291 0.0468 –0.0377 0.0409 –0.0352 –0.2343 –0.1514 1
Source: Authors’ own analysis.
Chinchwadkar and Seth 17

Type of Exit = a + R i b i (PE Investor Characteristics)

+ R j b j (Industry Factors)
+ R k b k (Market - timing Variables)
+ R l b l (Deal - specific Factors)
+ R m b m (demand - for - funds factors),

where the dependent variable is 1 if the exit method is an IPO and 0 if the
exit is via an acquisition. The details of the independent variables are
presented in Appendix A and are listed in the first column of Table 6.
We observe that all variables in the category ‘PE investor character-
istics’ have coefficients significantly different than 0. The ‘number of
PE investors’ has a positive coefficient in all three models at 1 per cent
level of significance. Thus, we understand that the existence of multiple
investors at the time of exit increases the probability of an IPO exit. This
effect is seen because the existence of multiple investors allows for better

Table 5. Univariate Difference Tests between IPO and Acquisition Samples

IPO Acquisition
Sample Sample Difference Parametric Wilcoxon
Variables Mean Mean in Means p-value p-value
PE Investor Characteristics:
Number of PE 1.9897 1.2675 0.7222*** <0.0001 <0.0001
Investors
Foreign 0.5816 0.4648 0.1168** <0.05 <0.05
Investor
Dummy
Co-investment 0.2653 0.0783 0.187*** <0.0001 <0.0001
Dummy
Industry Factors:
Financial 0.0714 0.0837 –0.0123 0.6913 0.6908
Services
Dummy
High-tech 0.3877 0.454 –0.0663 0.2407 0.2403
Dummy
Real Estate 0.0714 0.054 0.0174 0.5129 0.5123
Dummy
Deal Size 83.27 79.25 4.02 0.4299 <0.0001
Source: Authors’ own analysis.
Note: ***, **, * represent significance at 1 per cent, 5 per cent and 10 per cent level,
respectively.
18 Journal of Emerging Market Finance 17(1S)

Table 6. Multivariate Tests Using Logistic Regression

Dependent Variable: ‘Type of Exit’ (1: IPO, 0: M&A)
Independent Variables Model 1 Model 2 Model 3
Number of PE Investors (H1) 0.7608*** 0.9129***
Foreign Investor (H2a) –0.6204* –0.5241* –0.5237*
Co-investment (H2b) 0.7015 1.9314***
Financial Services (H3) –1.8993*** –1.7758** –1.5899**
High-tech (H4) –0.8399** –0.7976** –0.8822**
Real Estate (H5) –0.1798 –0.1451 –0.4002
Industry market-to-book (H6) 0.0901 0.0866 0.1223
Industry Herfindahl Index (H7) –0.0235 –0.1385 –0.05601
Log (Size of Deal) (H8) 0.9628*** 0.9483*** 1.0768***
Buyout Dummy (H9) –1.0850* –1.0888 –1.1773*
Recession Dummy (2009) –0.0355* –1.0484* –0.9440
(H10)
Market (t) 0.05559 –0.0357 –0.0348
Market (t – 1) –0.1141*** 0.0529*** 0.0535***
T-Bill Rate 0.76081 –0.1021 –0.1294
Source: Authors’ own analysis.
Note: ***, **, * represent significance at 1 per cent, 5 per cent and 10 per cent level,
respectively.

resolution of information asymmetry, thus making IPO exits easier. When

we analyse the effect of foreign investors, we observe that the presence of
foreign PE investors in the syndicate reduces the probability of an IPO exit
since foreign investors are not able to resolve information asymmetry as
well as domestic investors. We also observe that co-investment by foreign
and domestic investors increases the probability of an IPO exit. Thus,
we find results consistent with the theory that the type of exit depends
on how well the PE investors are able to resolve information asymmetry
issues with the buyers.
When it comes to industries, we find that financial services firms tend
to prefer acquisitions over IPOs as exit options. As studied by Berger
et al. (1999), this could be a result of many advantages of consolidation in
this industry. In high-tech firms, we find evidence supporting the theory
that high-tech firms prefer acquisitions over IPOs because they attract
significantly higher acquisition premium (N. Kohers & T. Kohers, 2000).
However, we find that in this simple logistic regression framework, indus-
try M/B ratio and industry concentration do not have a significant effect
on the exit decision.
When we analyse deal-specific factors, we come across a unique
finding. We find that the ‘type of entry’ has a significant effect on the type
Chinchwadkar and Seth 19

of exit. We see that if the type of entry is via a buyout, the exit outcome is
more likely to be an acquisition rather than an IPO. This result opens up
a new line of research to analyse the mechanism through which the type
of entry affects the exit choice. Next, the deal size comes across as an
extremely important variable for the exit choice. We observe that larger
the deal size, more is the probability of an IPO. This result stems from the
fact that higher fixed costs of an IPO discourage small firms from going
public (Holmstrom & Tirole, 1993; Ritter, 1987). Also, we find that the
recession year (2009) decreases the probability of an IPO exit and the
effect is significant over and above existing market conditions (measured
by market returns).
We find mixed evidence of the effect of market returns on the exit deci-
sion. Interestingly, we find that the 3-month T-Bill rate is not significant
for determining the type of exit, implying that the cost of debt is irrelevant
for the choice of exit between IPO and acquisition. This result is different
from the standard results in literature on developed markets like USA
and could stem from two features of the Indian market: (a) banks are not
allowed to finance acquisitions and (b) limited access to the corporate debt
market. The Indian market provides an opportunity to further examine the
relationship between the cost of debt and acquisitions.

5.2. Multinomial Logistic Regression Results

We now analyse the choice of exit in detail by considering four exit
methods. A nominal variable is created which has four categories: IPO,
financial sale, strategic sale and buyback. Since we cannot meaningfully
order these categories, we cannot use ordered-probit models. Hence, we
use multinomial logistic regressions as shown in Model (2):
Type of Exit (detailed) = a + R i b i (PE Investor Characteristics)
+ R j b j (Industry Factors)
+ R k b k (Market - timing Variables)
+ R l b l (Deal - specific Factors)
+ R m b m (Demand - for - funds Factors),
where the dependent variable takes the following values: 1 = IPO, 2 =
Financial sale, 3 = Strategic sale and 4 = Buyback. The details of the
independent variables are presented in Appendix A and are listed in the
first column of Table 7.
Multinomial logistic regressions help us delve deeper into the differ-
ent types of exit outcomes and analyse how each variable affects the exit
20 Journal of Emerging Market Finance 17(1S)

Table 7. Multivariate Tests Using Multinomial Logit Regression

Dependent Variable: ‘Type of Exit’ (1: IPO, 2: Financial Sale, 3: Strategic Sale,
4: Buyback)
Independent Variables 1 vs. 2 1 vs. 3 1 vs. 4
Number of PE Investors (H1) –1.947*** –0.3878*** –1.3299***
Foreign Investor (H2a) 0.4342 0.0649 1.162
Financial Services (H3) 2.9371*** 0.7861 2.5254**
High Tech (H4) 1.3284** 0.4477 0.6815
Real Estate (H5) 0.1163 –0.7409 1.9140**
Industry Market-to-book (H6) –0.1347 0.0532 –0.0547
Industry Herfindahl Index (H7) 3.5040 1.3735 –3.764
Log (Size of Deal) (H8) –0.9973*** –0.3645 –1.2423***
Buyout Dummy (H9) 1.3201 1.2748* 0.607
Recession Dummy (2009) (H10) 1.0376 0.4840 –0.8651
Market (t) 0.0465 0.0150 0.0829**
Market (t – 1) –0.0353 –0.0565** –0.0301
T-Bill Rate 0.1584 –0.0203 –0.1336
Source: Authors’ own analysis.
Note: ***, **, * represent significance at 1 per cent, 5 per cent and 10 per cent level,
respectively.

choice. This regression technique compares IPO to each of the other three
methods and reports how the independent variables affect the following
three choices: (a) IPO versus financial sale, (b) IPO versus strategic sale
and (c) IPO versus buyback.
We find that the variable ‘number of PE investors’ is significant and has
consistent signs for all three choices. At the same time, foreign investors
do not significantly affect this detailed level of decision-making. The pres-
ence of a firm in the financial services industry increases the probability of
financial sales and buybacks, whereas firms in the high-tech industries will
prefer financial sales as compared to IPOs. We come across another unique
finding concerning the real estate industry. We find that presence in the
real estate industry significantly increases the probability of buybacks as
compared to other exit outcomes. More or less, we find consistent results
for other variables as well and observe that buyouts affect strategic sales
more than other exit outcomes.

5.3. Analysis of Marginal Effects

The standard estimates of a multinomial logistic regression assume a
base case (in our case it is IPO) and then present results by comparing all
Chinchwadkar and Seth 21

the options to the base case. Thus, all the analysis has been done by com-
paring financial sale, strategic sale and buyback to IPOs. However, it
would be insightful to study the effect of independent variables on all the
exit outcomes at the same time. For this purpose, we estimate marginal
effects of the multinomial logistic regression which allow us to study the
effects of independent variables on all exit methods simultaneously
(Table 8). Although the results of the marginal effects analysis are con-
sistent with earlier findings, we gain some incremental information by
studying the results.
We observe that the size of the syndicate (number of PE investors)
and financial services dummy are significant factors for all exit outcomes
except buybacks. The presence of foreign investors in the syndicate

Table 8. Marginal Effects of Multinomial Logit Regression

Dependent Variable: ‘Type of Exit’ (1: IPO, 2: Financial Sale, 3: Strategic Sale, 4:
Buyback)
Independent
Variables IPO Financial Sale Strategic Sale Buyback
Number of PE 0.2989*** –0.1078** 0.1645*** –0.0363
Investors (H1)
Foreign Investor –0.1075 0.0029 –0.0919 0.0764*
(H2a)
Financial Services –0.2901*** 0.2248* –0.1813** 0.1121
(H3)
High Tech (H4) –0.1791** 0.0800 –0.0465 0.0054
Real Estate (H5) –0.0598 –0.0101 –0.1826** 0.4069**
Industry Market- 0.0221 –0.0083 0.0339* 0.00048
to-book (H6)
Industry Herfindahl 0.1905 0.4497** 0.5550 –0.3299
Index (H7)
Log (Size of Deal) 0.1862*** –0.0412 0.0738 –0.0629**
(H8)
Buyout Dummy –0.1631* 0.0938 0.2238 –0.0073
(H9)
Recession Dummy –0.1421 0.0726 –0.0089 –0.0910***
(2009) (H10)
Market (t) –0.0086* 0.0022 –0.0034 0.0058*
Market (t – 1) 0.01087** –0.00017 –0.0066 0.00033
T-Bill Rate –0.0150 0.0149 –0.0148 –0.0159
Source: Authors’ own analysis.
Note: ***, **, * represent significance at 1 per cent, 5 per cent and 10 per cent level,
respectively.
22 Journal of Emerging Market Finance 17(1S)

matters only in case of buybacks. Apart from this, we see that the deal size
matters only for IPOs and buybacks; large firms prefer IPOs and avoid
buybacks. When it comes to market conditions, previous month market
return seems to have a positive effect on IPO exits. We find that IPOs
tend to follow months in which the market return was high, indicating a
market-timing element to the exit choice. These findings provide a base
for interesting new research into the mechanisms driving these results.

6. Conclusions
PE investors and entrepreneurs face a fundamental and inevitable
decision to choose between exit methods: IPO or acquisition. Previous
studies have highlighted the importance of four categories of variables
affecting this choice: industry features, deal-specific factors, market-
timing variables and demand-for-funds factors in this choice. However,
to the best of our knowledge, there has been little research on the effect
of ‘PE investor characteristics’ on this choice. Using a sample of all PE
exits in India between 2004 and 2010 (the first PE cycle in India), we
address this question by examining the effect of three PE characteristics
on the choice of exit: size of PE syndicate, presence of foreign PE inves-
tors and co-investment by foreign and domestic PE investors. Our results
show that a larger PE syndicate increases the probability of an exit through
IPO. Also, the presence of foreign PE investors and co-investment by
foreign and domestic PE investors increases the probability that the PE
investors will exit via an IPO rather than an acquisition. We further ana-
lyse the detailed exit outcomes such as financial sale, strategic sale and
buyback to find that PE investors’ characteristics significantly affect
each of the detailed outcomes. Our results are consistent with the theory
that information asymmetry resolution is a dominant driver of exit out-
comes. Overall, our study sheds light on a previously unexplored dimen-
sion of the choice of exit between IPO and acquisition, identifies key
characteristics of PE investors which affect this decision and presents
interesting new results which can be a foundation for future research.
Appendix A: Description of Variables

Variable Description Source

A. Independent Variables:
Type of Exit Binary variable. 1: IPO, 0: Acquisition Venture Intelligence
Type of Exit (detailed) Categorical variable. 1: IPO, 2: Financial Sale, 3: Strategic Sale, 4: Buyback Venture Intelligence
B. PE Investor Characteristics:
Number of PE Investors This variable represents the total number of investors exiting from a firm Venture Intelligence
Foreign Dummy Binary variable. 1: Foreign PE investor has invested in the firm Venture Intelligence
Co-investment Dummy Binary variable. 1: Domestic and Foreign PE investors have invested together in the firm Venture Intelligence
C. Industry-related Factors:
Financial Services Dummy Binary variable. 1: Firm is in the financial services industry CMIE Prowess
High-tech Dummy Binary variable. 1: Firm is in a technology related industry CMIE Prowess
Real Estate Dummy Binary variable. 1: Firm is in the real estate sector CMIE Prowess
Industry Market-to-book This variable represents the average market-to-book ratio of the industry of the exiting firm CMIE Prowess
Industry Herfindahl Index This variable represents the Herfindahl Index (concentration ratio) of the industry of the CMIE Prowess
exiting firm
D. Deal-specific Factors:
Deal Size This is the total size of the exit in US$ million Venture Intelligence
Buyout Dummy Binary variable. 1: The entry by PE investors has been through a buyout Venture Intelligence
Recession Dummy Binary variable. 1: The exit was executed in 2009 (recession year) Venture Intelligence
E. Market-timing Variables:
Market (t) This variable represents the market return (BSE SENSEX) in the month of the exit BSE Archives
Market (t – 1) This variable represents the market return (BSE SENSEX) in the month previous to the exit BSE Archives
F. Demand-for-funds Factors:
T-Bill This variable represents the 3-month T-Bill rate (India) in the month of the exit Bloomberg
24 Journal of Emerging Market Finance 17(1S)

References
Annamalai, T., & Doshi, M. (2012). Beyond capital: PE and real estate
development in India. Journal of Private Equity, 15(3), 62–76.
Bayar, O., & Chemmanur, T. J. (2012). IPOs versus acquisitions and the valuation
premium puzzle: A theory of exit choice by entrepreneurs and venture capi-
talists. Journal of Financial and Quantitative Analysis, 46(6), 1755–1793.
Benninga, S., Helmantel, M., & Sarig, O. (2005). The timing of initial public
offerings. Journal of Financial Economics, 75(1), 115–132.
Berger, A., Demsetz, R., & Strahan, P. (1999). The consolidation of the financial
services industry: Causes, consequences, and implications for the future.
Journal of Banking and Finance, 23(2), 135–194.
Black, B., & Gilson, R. (1998). Venture capital and the structure of capital
markets: Banks versus stock markets. Journal of Financial Economics,
47(3), 243–277.
Brau, J. C., Francis, B., & Kohers, N. (2003). The choice of IPO versus takeover:
Empirical evidence. Journal of Business, 76(4), 583–612.
Chemmanur, T. J., & Fulghieri, P. (1999). A theory of the going-public decision.
Review of Financial Studies, 12(2), 249–279.
Cochrane, J. (2005). The risk and return of venture capital. Journal of Financial
Economics, 75(1), 3–52.
Cumming, D., & Macintosh, J. (2003). A cross-country comparison of full and
partial venture capital exits. Journal of Banking & Finance, 27(3), 511–548.
Cumming, D., & Johan, S. (2006). Provincial preferences in PE. Financial
Markets and Portfolio Management, 20(4), 369–398.
———. (2008). Information asymmetries, agency costs and venture capital exit
outcomes. Venture Capital, 10(3), 197–231.
Farag, H., Hommel, U., Witt, P., & Wright, M. (2004). Contracting, monitoring,
and exiting venture investments in transitioning economies: A comparative
analysis of Eastern European and German markets. Venture Capital,  6(4),
257–282.
Fleming, G. (2004). Venture capital returns in Australia. Venture Capital, 6(1),
23–45.
Golbe, D. L., & White, L. J. (1993). Catch a wave: Time series behavior of
mergers. Review of Economics and Statistics, 75(3), 493–500.
Gompers, P. A. (1996). Grandstanding in the venture capital industry. Journal of
Financial Economics, 42(1), 133–156.
Helwege, J., & Liang, N. (2004). Initial public offerings in hot and cold
markets. Journal of Financial and Quantitative Analysis, 39(3), 541–569.
Holmstrom, B., & Tirole, J. (1993). Market liquidity and performance monitor-
ing. Journal of Political Economy, 101(4), 678–709.
Kohers, N., & Kohers, T. (2000). The value creation potential of high tech
mergers. Financial Analysts Journal, 56(3), 40–50.
Korajczyk, R. A., Lucas, D. J., & McDonald, R. L. (1992). Equity issues with
time-varying asymmetric information. Journal of Financial and Quantitative
Analysis, 27(3), 397–417.
Chinchwadkar and Seth 25

KPMG Report. (2011). Returns from Indian Private Equity: Will the industry
deliver to expectations? India: KPMG. Retrieved from https://www.kpmg.
de/docs/Return-from-Indian-Private-Equity_1.pdf
Kutsuna, K., Cowling, M., & Westhead, P. (2000). The short-run performance of
JASDAQ companies and venture capital involvement before and after flota-
tion. Venture Capital, 2(1), 1–25.
Lowry, M., & Schwert, G. W. (2002). IPO market cycles: Bubbles or sequential
learning? The Journal of Finance, 57(3), 1171–1200.
Maksimovic, V., & Pichler, P. (2001). Technological innovation and initial public
offerings. Review of Financial Studies, 14(2), 459–494.
McKaskill, T., Weaver, K. M., & Dickson, P. (2004). Developing an exit readi-
ness index: A research note. Venture Capital, 6(2/3), 173–179.
Myers, S., & Majluf, N. (1984). Corporate financing and investment decisions
when firms have information that investors do not have. Journal of Financial
Economics, 13(2), 187–221.
Neus, W., & Walz, U. (2005). Exit timing of venture capitalists in the course of an
initial public offering. Journal of Financial Intermediation, 14(2), 253–277.
Nikoskelainen, E., & Wright, M. (2007). The impact of corporate governance
methods on value increase in leveraged buyouts. Journal of Corporate
Finance, 13(4), 511–537.
Pagano, M., & Roell, A. (1998). The choice of stock ownership structure:
Agency costs, monitoring, and the decision to go public. Quarterly Journal
of Economics, 113(1), 187–225.
Parhankangas, A., Landström, H., & Smith, D. G. (2005). Experience, contractual
covenants and venture capitalists’ responses to unmet expectations. Venture
Capital, 7(4), 297–318.
Pástor, Ľ., & Veronesi, P. (2005). Rational IPO waves. The Journal of Finance, 60(4),
1713–1757.
Poulsen, A. B., & Stegemoller, M. (2008). Moving from private to public own-
ership: Selling out to public firms versus initial public offerings. Financial
Management, 37(1), 81–101.
Povaly, S. (2007). PE exits: Divestment process management for leveraged buy-
outs. Germany: Springer.
Ritter, J. R. (1987). The costs of going public. Journal of Financial Economics,
19(2), 269–281.
Ritter, J. R., & Welch, I. (2002). A review of IPO activity, pricing, and alloca-
tions. The Journal of Finance, 57(4), 1795–1828.
Schultz, P. (2003). Pseudo market timing and the long-run underperformance of
IPOs. Journal of Finance, 58(2), 483–517.
Schultz, P., & Zaman, M. (2001). Do the individuals closest to internet firms
believe they are overvalued. Journal of Financial Economics, 59(3), 347–381.
Sharma, A., & Kesner, I. F. (1996). Diversifying entry: Some ex ante explanations
for postentry survival and growth. Academy of Management Journal, 39(3),
635–677.
26 Journal of Emerging Market Finance 17(1S)

Subrahmanyam, A., & Titman, S. (1999). The going-public decision and the
development of financial markets. The Journal of Finance, 54(3), 1045–
1082.
Yosha, O. (1995). Information disclosure costs and the choice of financing
source. Journal of Financial Intermediation, 4(1), 3–20.
Zingales, L. (1995). Insider ownership and the decision to go public. The Review
of Economic Studies, 62(3), 425–448.