Antecedents of Stage-wise Investment Preferences of Venture Capital and Private Equity Firms in India: An Empirical Exploration

1. Introduction

Venture capital and private equity (VCPE) has emerged as a new asset class to make funding available to entrepreneurs with innovative business ideas having high growth prospects, but facing difficulty in procuring finance as they are subject to high levels of information asymmetry and risk (Reddy, 2009). VCPE firms have evolved as a potential source of venture finance in developed and developing economies, supplementing traditional sources of resource mobilization such as public equity issues, private placements, and external commercial borrowings (Jain & Manna, 2009).

Entrepreneurial ventures across sectors of an economy, with varying technology and market risks, require funding at various stages of their development cycle (Robinson, 1987). Initially, VCPE firms provided only early-stage financing to technology-driven enterprises with a large growth potential. Over time, these firms also recognized the need to finance subsequent stages like expansion, growth, and buyout stages of enterprises when they are not able to raise funds through traditional sources (Lam, 1991). Today, VCPE firms are investment vehicles that invest at different stages (early to late) of an investee firm’s life cycle (Gultekin, 2010). Investments made by these firms across different stages are subject to varying levels of risks in an uncertain environment wherein an entrepreneur faces an imperfect capital market and an investor faces problems of moral hazard, asymmetric information, and adverse selection (Akerlof, 1978; Amit et al., 1998; Wang & Zhou, 2004). Consequently, VCPE firms have developed their own criteria to identify the stage of investment in order to manage the risks and ensure optimum returns (Carter & Van Auken, 1994). A VCPE firm may decide to invest in different stages (early or late) of an enterprise, but the criteria for decision-making at each stage may be different based on the time scale and the risk levels involved. Typically, firms would expect higher rates of return in early-stage investments due to the higher risk as compared to the late-stage investments (Lerner, 1994; Schilit, 1997).

Studies in the developed and emerging economies on investment strategies of VCPE firms have identified adoption of stage-wise investing by these firms as a means for risk-return trade-offs, diversification or specialization strategies, or simply a mechanism to earn better returns (Carter & Van Auken, 1994; Gompers, 1995; Gupta & Sapienza, 1992; Norton & Tenenbaum, 1993). Earlier studies also suggest that the risk-return assessment of VCPE firms while investing across stages is likely to be affected by factors like inherent risk of the venture, size of the investment, type of industry, and the physical location of the investee company (Carter & Van Auken, 1994; Gompers & Lerner, 2001). However, very few studies have explored the determinants of investment by stages and there exists no such study in the context of emerging economies like India. Moreover, extant studies have not analyzed the role of VCPE firm and deal characteristics in influencing the stage of investments. This article is the first to fill this research gap by examining the role of these characteristics while controlling for other variables that have been identified to influence VCPE investments which include industry and geographical location of the enterprise apart from the size of investment.

The focus on India in this article is also very relevant as it is one of the fastest growing emerging economy in the world and has seen VCPE investments grow rapidly from US$152 mn in 1998 to US$ 14,568 mn in 2016 with a CAGR of 29%. It has been among the top 10 out of 77 nations in the world in terms of VCPE investments (Rajan & Deshmukh, 2011). These investments have been made across industries and locations and during the last two decades ending 2016, while 27% were early-stage investments, the remaining were late-stage ones (http://www.ventureintelligence.com). Given the diversity of investments in terms of industry, region, and stage preference of VCPE firms, India provides an appropriate emerging country context to explore the determinants of stage-specific investment strategies of investing firms. This is done by analyzing a total of 5,782 investment deals made by VCPE firms in India during 1998–2016.

The rest of this article is organized as follows. Section 2 provides a brief review of literature and develops tentative hypothetical relationships. Section 3 gives details of the data and methodology. Section 4 presents the empirical results. Section 5 discusses the results and the final section concludes.

2. Literature Review

At the core of VCPE investments is the management of information asymmetry between the investor and investee. Typically, the entrepreneur knows more about a project than the financier and it is difficult to measure and monitor inputs of the entrepreneur. This results in standard principal–agent problem that the investor and the investee need to resolve (Akerlof, 1978; Amit et al., 1990; Sahlman, 1990). The information asymmetry and the associated agency problems are likely to be higher when the projects are based on new innovations and in the early stages of an entrepreneurial life cycle as both technological and market uncertainty are high in such situations (Lerner, 1994). Standard project evaluation procedures are inappropriate in such cases and usually specialized intermediaries like early-stage investors undertake such investments. Often these investors have their own idiosyncratic heuristics to evaluate projects (Kerr & Nanda, 2015). Apart from monitoring, board representation, and replacing founders with professional CEOs in underperforming ventures, stage-wise investing is an important mechanism to take care of information asymmetry-driven agency problems and also reduce risks (Chemmanur et al., 2011; Kerr & Nanda, 2015). Moreover, the heuristics of evaluating a project may also change in different stages as more information is available in later stages of an entrepreneurial life cycle. This is consistent with the studies that show that evaluation frameworks used by business angels differ from those used by venture capitalists. In fact, it is also observed that business angels and venture capitalists rank and assess investment criteria like management team, business potential, and financial traction quite differently for a project (Granz et al., 2020).

It is virtually impossible to get large-scale data on different variables that capture in detail the team composition, business conditions, and market/financial traction for different deals that have been funded by VCPE firms. Therefore, analyzing the reasons for investing at different stages using detailed deal or firm-specific variables is not possible. Studies have tried to infer conditions of information asymmetry, riskiness of a project, risk aversion or specialization of investors, and differences in heuristics from data on characteristics of investors, investees, and investment deals (Carter & Van Auken, 1994; Gompers, 1995; Gupta & Sapienza, 1992; Norton & Tenenbaum, 1993). We pool together below some insights from the literature on how certain characteristics of VCPE firms, investee firms, and investment deals can influence stage-wise investment strategies of investor firms.

3. Data and Methodology

The analysis is based on 6,462 VCPE deals executed from April 1998 to March 2016 in 3,841 companies which are sourced from Venture Intelligence Database. Out of these, investment data for 5,782 deals was available, with an aggregate investment of US$117854 mn. This might not include all of the VCPE deals during the period, since many of the deals may not have been announced, but we believe the data captures a majority of the investments that happened during the period, and more importantly is representative of the industry trends. The data is available at the deal level which is self-reported by the VCPE firms and used extensively by various stakeholders.

We first explore the data on VCPE firms’ investment preferences in various stages and their correlates through cross tabulation and chi-square tests. This is followed by binary logistic regression analysis to test the impact of individual variables on the stage-specific investment preferences of VCPE firms in India.

3.1. Variable and Model Description

The database provides information for each VCPE deal including the name, industry, and location of the investee company; investment amount; stage of investment; name and type of investor(s); and deal date. In our analysis, each deal is taken as an independent decision of the VCPE firm to invest. Thus, two rounds of funding in the same investee firm are treated as two different investment commitments to the same investee firm.

Table 1 presents details of the variables used in our analysis. Stage of investments made by VCPE firms are classified in seven subcategories based on the life cycle stage of the investee firm, the duration, and amount of the investment: early (<5 years); growth (<10 years); late (>10 years); pre-IPO; private investments in public equity in listed companies (PIPE); buyout (acquisition of controlling or significant stake in investee firm); and others. For the regression analysis, age of the VCPE firm at the time of the deal is used as a proxy for firm experience. For chi-square tests, VCPE firms with more than median age are classified as experienced firms and less than or equal to the median age are classified as inexperienced. The type of VCPE investors in a deal are classified as India-dedicated investors—who make investments in India from India-dedicated VCPE funds, foreign investors—who invest in India out of overseas funds, and co-investment indicates investments made by both India-dedicated and foreign investors. Type of industry of the investee firm for each deal is taken from the database and reclassified as per the classification used by IVCA Report, 2009 into nine categories. Based on the geographical location of their corporate headquarters, investee firms are classified into six regions. Presence of more than one investor in a VCPE deal is classified as a syndicated deal; a non-syndicated deal has only one investor. Size of investment of each deal is in US$ mn. For chi-square tests, based on the quartiles, investments were classified into three categories: small, medium, and large.

OPEN IN VIEWER

Table 1.

Description of Variables

Variable	Details	Description
VCPE firm experience	Experienced Inexperienced	Age is taken as the proxy for experience and measured as the difference between date of deal and the date of incorporation of the firm.
Type of investor	India-dedicated * Foreign Co-investment	*Reference category Dummies are used for foreign investors and co-investment.
Type of deal	Non-syndicated* Syndicated	*Reference category Dummy for syndicated deals
Size of investment	Deal size	Log of amount of investment in a deal in US $ mn
Type of industry	Telecom and media (T&M)* Information & technology (IT&ITES) Banking, financial services and insurance (BFSI) Manufacturing (MFG) Healthcare (HC) Engineering and construction (E&C) Shipping and logistics (S&L) Non-financial services (NFS) Others (OT)	*Reference category A dummy is used for each industry.
Region of investee firm	West* South North East Central Overseas	*Reference category A dummy is used for each regional location for the investee firm
Stage of investment	Early Growth Late PIPE Pre-IPO Buyout Others	1 if the stage of investment is early 0 if the stage of investment is late For our empirical analysis, the first two stages (early and growth) are categorized as early stage as they capture VCPE investments in firms that are less than 10 years old. All subsequent stages are categorized as late-stage investments undertaken in firms more than 10 years old.

Source: Compiled from Venture Intelligence Database, India.

Note: * indicates the reference category of a variable where there are dummy variables.

We use the binary logistic regression for our regression analysis as it is considered to be one of the best methods where the outcome is in the form of happening or nonhappening of an event (Yarandi & Simpson, 1991). This method was best suited for the current study as the dependent variable is a dichotomous categorical variable and the explanatory variables are both categorical and continuous in nature. The seven stages of investment as described above are reclassified into early and late (see Table 1) to have a dichotomous-dependent variable representing the outcome as a likelihood of a VCPE investment deal happening in the early or late stage of the investee firm’s life cycle. In our analysis, six variables (across VCPE firm, investee firm, and deal characteristics) are used to predict the one dichotomous outcome variable. The variables include VCPE firm experience, type of VCPE investor, and type of deal which are our key explanatory variables. Type of industry, regional location of investee firm, and size of investment deal are the control variables. Except for VCPE experience and size of investment, all other explanatory variables are categorical.

The following model has been estimated:

S t a g e o f i n v e s t m e n t ( Y ) = β 0 + β 1 ( V C P E f i r m exp ​ e r i e n c e ) + β 2 ( i n v e s t o r t y p e ) + β 3 ( d e a l t y p e ) + β 4 ( d e a l s i z e ) + β 5 ( i n v e s t e e f i r min d u s t r y ) + β 6 ( i n v e s t e e f i r m r e g i o n ) + ε i

(1)

The dichotomous-dependent variable, stage of investment (Y), takes the value equal to 1 if the stage of investment was early and equal to 0 if the stage of investment was late. ß₀ is the Y intercept and ß₁ to ß₆ are the beta coefficients for the various explanatory and control variables. Table 1 provides details of how each variable is measured. Investor types distinguish foreign and domestic ownership. Nine industry groups and six regions have been identified and deals are categorized into syndicated and non-syndicated ones. Dummies are used for each industry, region, and deal type. εi is the error term with logistic distribution.

As discussed, two types of empirical analyses have been undertaken. We begin with a simple description of data which is followed by chi-square tests and regression analysis.

The empirical results discussed in the last two sections provide useful insights on the determinants of early-stage investments in India. As discussed, with experience, VCPE firms may prefer early-stage investments as they have learnt to deal with the risks and issues relating to information asymmetry by developing the relevant protocols and heuristics. Our results suggest otherwise as experienced VCPE firms prefer to invest in late stages. One potential explanation is that older VCPE firms in India have ended up in specializing in late-stage investments as they may not require exposure to early-stage deals to establish themselves. And younger firms are opting for early-stage investments to build reputation as they experiment with their own idiosyncratic heuristic models or in anticipation of higher returns as suggested by some studies (Gompers, 1995; Schwienbacher, 2008). Apparently, experience of VCPE firms in India, as suggested by many other studies, is being leveraged by them to benefit from specialization in investing in a particular stage which gives them an edge over others in terms of managing risks and earning better returns (Carter & Van Auken, 1994; Gompers et al., 2008; Gupta & Sapienza, 1992). Since we did not have the relevant data, we are not able to control for fund size. In all likelihood, older VCPE firms would have larger investible funds and therefore an ability to spread their investments across various stages of the entrepreneurial life cycle. Younger VCPE firms with their smaller fund size may not be able to make large number of investments across stages and end up focusing more on early-stage deals.

The preference of VCPE firms with foreign investors for late-stage investments vis-à-vis India dedicated investors may partly be driven by information asymmetry associated with early-stage deals, which they are not able to manage due to lack of geographical proximity to the investee firms (Annamalai & Kamat, 2012). Conversely, the geographical proximity of VCPE firms with India-dedicated investors to investee firms helps them monitor progress and manage such risks better resulting in higher preference for early-stage investments (Joshi & Subrahmanya, 2015; Pruthi et al., 2003). Besides, they are likely to have a better understanding of the Indian context.

Syndicated deals being more likely to be early-stage deals as compared to non-syndicated ones, which are more likely to be late-stage investments, suggest that syndication is being used to manage higher risks of early-stage investments, share expertise, and resources. This results in better selection and thus reduces the risk of VCPE firms to invest in early stages and this pattern is similar to that observed elsewhere (Casamatta & Haritchabalet, 2007; Deli & Santhanakrishnan, 2010; Hopp & Rieder, 2005; Vu & Mireille, 2011).

An increase in deal size reduces the likelihood of the deal being an early-stage one. This is probable because in early stages, the funding requirements are lower and as the company progresses and plans for expansion of its scale of operations, the funding needs increase, requiring larger amounts of investments from its investors (Dhochak & Sharma, 2015; Gemson & Annamalai, 2015). Moreover, in situations of high information asymmetry and uncertainty, which characterize early-stage investing, one would not like to make large investments. Thus, Indian VCPE firms seem more likely to invest in early stages where the deal size is small and if the deal size is large, it is more likely to be a late-stage deal. Since early-stage investments are usually small, this can be seen as intuitive and somewhat tautological but since we are controlling for region and industry, this result seems consistent with expectations.

As mentioned, specialization in specific industries can help VCPE firms to develop superior skills of selecting and monitoring entrepreneurial projects in that industry by building their own idiosyncratic heuristics. Besides, information asymmetry and risks involved in different stages of investments vary by industry as industries differ in terms of lock-in periods, exit opportunities, market conditions, and intrinsic riskiness. Also, if start-up activity is particularly high in an industry, it will throw up more early-stage investment opportunities as compared to sectors where new enterprise creation is not so dominant. We find that Indian VCPE firms are more likely to invest in early stages in the IT&ITES industry. It is probably because the start-up activity in India is dominated by this sector and there is a preponderance of early-stage deals in this industry. Also, enterprises in this sector have shorter entrepreneurial lifecycles, high innovation quotient, and lower investment requirements, which are more promising in terms of future returns. Often, start-ups in this sector have strong revenue growth potential but negative cash flows and thus attract angel or VCs who invest in early stages rather than private equity funding where the focus is on growth companies. Within the remaining sectors, industries that are preferred for early-stage investments are nonfinancial services, telecom, and media and health care.

Overseas, South and North regions are more likely to attract early-stage investment as compared to East and Central regions. This could be because VCPE firms that invest in the North and South regions are positioned closer to their investee firms which makes it possible for them to assess and monitor the risks of early-stage investing and help their investee firms grow (Florida & Smith, 1993). Moreover, the ecosystems for early-stage ventures is more mature in the North and South regions and therefore can support the growth of start-ups located in these regions. Absence of these features in East and Central regions makes VCPE firms to prefer safer late-stage investments.

The presence of a vibrant entrepreneurial ecosystem, concentration of IT&ITES industries facilitating high start-up activity, and geographical proximity to the investee firms have influenced the VCPE firms to take risks by investing in early stages if the investee firm was located in the North and South regions of the country. While the West region has a concentration of BFSI industries wherein start-up activity is not very high, it attracts more late-stage investments. Thus, concentration of VCPE firms facilitating ability to a monitor information asymmetry issues, higher levels of start-up activity and a robust ecosystem in a region provide higher opportunities for early-stage investments.

6. Some Concluding Observations

For almost three decades now, VCPE investments have contributed to Indian entrepreneurial ventures, characterized by varying degrees of innovation and market risks. Extant literature suggests that a key issue in VCPE investments is the management of information asymmetry between the investor and the investee which is likely to be higher in the early stages of the ventures. VCPE firms build specialized skills and idiosyncratic heuristics to evaluate early-stage vs. late-stage investments which are different from conventional project evaluation methods and also differ across stages. This article analyzed 5,782 VCPE investment deals from 1998 to 2016 to ascertain if factors like VCPE firm experience, type of investor, type of industry, location of the investee firm, type of deal being syndicated or not, and the size of investment impact stage-specific investment decisions of VCPE firms. To the best of our knowledge, no empirical study in India has been done in this regard.

With extreme levels of information asymmetry attached to VCPE funding, firms engaged in such financing need to possess specialized risk and assessment skills to make the right investments. Our empirical analysis suggests that VCPE firms in India that invest in early stages of the entrepreneurial life cycle focus on specific industries that have projects with shorter gestation periods and lower capital intensity so that the investments are not high and the payback periods are short. Lower capital intensity also reduces the risk of not being able to raise follow-on capital and creates exit possibilities, making the risk of early-stage investment somewhat lower. They choose those locations for such investments where there is a cluster of such enterprises and various stakeholders of the ecosystem are also present to actively supplement the efforts of the VCPE firms to make the investee firm grow. These results are consistent with earlier studies on this subject.

Our contribution is more in the exploration of the impact of VCPE firm and deal characteristics on the stage of investment. We find that syndication is actively used to reduce risks and uncertainties when early-stage investments are made. Domestic VCPE firms prefer early stage investments vis-à-vis their foreign counterparts due to their superior ability to understand domestic business context and undertake monitoring, and so on to manage information asymmetry in such investments. Interestingly, experience (age) of the investor firm does not result in higher risk-taking; in fact, younger firms have a higher propensity to make early-stage investments. But it is possible that age of a firm may not fully reflect experience as senior management of a young firm can also have a lot of experience and having the right heuristics to deal with early-stage investments.

While the article provides useful insights about choices relating to stage of investment in an emerging market context of India, there are limitations which open up avenues for future research. Secondary data used in this study is self-reported by VCPE firms and may be incomplete or contain errors. Due to nonavailability of data, the role of VCPE fund size could not be explored. Each deal was treated as an independent investment decision and we could not explore the linkages between early- and late-stage investments made by the same VCPE firm due to the absence of panel data. With better data availability, these gaps can be addressed. For an initial exploration, this article only categorized deals into early and late stage ones, although the data defines seven investment categories. One can potentially undertake a multinomial logit exercise to explore the antecedents of investments in multiple stages. Using all seven stages may not be very meaningful, but distinguishing among early, growth, and late stages can add useful insights. Such an exercise, would, of course, entail a more nuanced reading of the extant literature to identify the role of factors analyzed in this article (as well as others) in affecting investment choices at multiple stages. Finally, VCPE financing being a global phenomenon, a comparative understanding of the stage-wise investment strategies followed by VCPE firms across countries with different contexts might be rewarding and generate newer insights.

Apart from contributing to the general literature on entrepreneurship, this study provides useful insights on VCPE financing in the emerging economy context, especially on the factors that affect stage-specific investment choices of the investing firms. We hope that our study can lead to a deeper investigation of the process of stage-wise investment decision-making of VCPE firms.