Can Spending Time in the Minors Pay Off? An Examination of the Canadian Junior Public Equity Markets

Abstract

This paper tests whether the junior public equity markets serve as an effective development market for early‐stage firms compared to private venture capital (VC). Focusing on the Canadian market, we compare the long‐run stock performance of firms that graduate from the Toronto Venture Stock Exchange (TSX‐V) to the Toronto Stock Exchange (TSX) against the performance of VC‐backed firms that have a direct IPO on the TSX. Controlling for various confounders and possible selection biases, we find that TSX‐V graduations significantly outperform VC‐backed IPOs by 28.2 percentage points in the three years following the TSX listing. Our results are consistent with the idea that a TSX‐V listing provides the insiders of early‐stage companies with valuable public‐markets experience.

Notes

1 Our sample of VC‐backed financings includes investments in non‐technology firms, which comprise both traditional VC investors and other PE investors. For the purposes of our study, the distinction between VC and PE investors is not considered relevant, and thus for ease of exposition we call all such investors VC investors in the paper.

2 See Carpentier and Suret (2010), which refers to the TSX‐V as a public VC market in the title of the article. A simple internet search also yields several citations of the TSX‐V as a public VC market.

3 See Metrick and Yasuda (2011) for excellent survey articles.

4 See Table 18 in “Initial Public Offerings: Updated Statistics on Long‐run Performance” on Professor Jay Ritter's website (https://site.warrington.ufl.edu/ritter/files/2017/03/IPOs2016Longrun.pdf).

5 The concept of “learning by doing” has a long history. Arrow (1962) was the first to apply the concept to the study of firms. Since the economics literature on learning‐by‐doing is too large to review here, we refer the reader to available surveys, such as by Thompson (2010).

6 Interestingly, both VC‐backed IPOs and TSX‐V graduations to the TSX were not particularly affected by the major capital market disruptions over this period.

7 See, for example, Ritter (1991) and Loughran and Ritter (1995) for U.S. evidence and Jog (1997) and Kooli and Suret (2004) for Canadian evidence.

8 Carpentier, L'Her, and Suret (2010) include alternative public listings such as RMs and Capital Pool Companies (CPCs) in their sample. We exclude these in our sample because they are not regular IPOs of operating companies. In an RM, a private company acquires a company that is already listed on the stock market, but effectively is only an empty shell company. As Carpentier, Cumming, and Suret (2012) document, RMs provide less disclosure to investors than IPOs, suffer from a higher degree of information asymmetry between the firm and its investors, and have poor performance compared with regular IPOs. A CPC is a specialized form of a “blind pool” offering that is regulated by the stock exchange. As noted in Carpentier, L'Her, and Suret (2010) and Pandes and Robinson (2013), a large portion of the TSX‐V IPOs are CPCs (see also Pandes and Robinson 2014; Robinson 1997).

9 Robinson and Cottrell (2007) also document that a significant amount of PE capital in the Canadian province of Alberta comes from informal investors as opposed to formal (VC) investors.

10 The main example of a European exchange‐regulated market is the LSE's AIM. Gerakos, Lang, and Maffett (2013) study firms listed on the AIM and show that its newly listed firms underperform firms that are listed on more established exchanges with higher regulations. Jenkinson and Ramadorai (2013) examine firms that switch between the AIM and main market over the period 1995–2006 and find that prior to the end of 2000, firms would predominantly switch from the junior to the senior exchange, but since then firms have mostly been switching from the senior to the junior exchange.

11 For example, the TSX‐V firms have less onerous requirements for board member independence than TSX firms.

12 Our sample of TSX‐V graduation firms is restricted to regular IPOs so that they are more directly comparable to VC‐backed TSX IPO firms. As such, we exclude TSX‐V firms created by a RM and firms created by a CPC IPO.

13 The FP database is a new issues database in Canada. It is equivalent to the Securities Data Corporation (SDC) new issues database in the United States.

14 Tunali (1986) shows that, after jointly estimating equations (2) and (3) through a bivariate probit model, two selectivity instruments, and , which are the direct analogues of the Inverse Mills Ratio proposed by Heckman in the single selection context, can be computed and included in the estimation of equation (1) to correct for the potential biases in the estimation of the parameters of interest (i.e., and the other coefficients). In our regression estimates of equation (1), we therefore include the two selectivity instruments, identified as IMR_TSX‐V and IMR_TSX. By doing so, the formulation in our paper is a specific case of the bivariate probit model, denoted a recursive model, since the endogenous variable [TSX‐V] appears on the right‐hand side of equation (3). The coefficient δ captures the effect of a TSX‐V listing on the probability of listing on the TSX, while accounting for the correlation between the two selection equations. If the correlation between the error terms of the two rules, , is equal to zero, the terms reduce to two independent correction terms, as in Heckman's standard method. On the other hand, if , neglecting the statistical relationship between the selection rules would lead to inconsistent estimates.

15 The estimated marginal effect implies a 4.5% decrease per year in the probability of going public on the TSX‐V.

16 The TSX/CFMRC value‐weighted index is the market value weighted average monthly return for all domestic common equities in the Canadian Financial Markets Research Centre (CFMRC) database. This index is the Canadian equivalent of the CRSP value‐weighted index in the United States.

17 The average leverage ratio of VC‐backed IPOs is especially high due to some large outliers. The median leverage ratio of this group is 39.5%. In the sub‐sample of graduations, the median leverage ratio is 18%. As a term of comparison, Michel (2014) finds that the average leverage ratio of U.S. VC‐backed IPO firms is 25.7%.

18 In order to control for potential multicollinearity concerns, we compute Variance Inflation Factors (VIF) for all BHAR regressions in the paper. Using an approach recommended by O'Brien (2007) to account for our small sample size, we find the average VIF to be lower than 2.5 in all our regressions and the individual‐variable VIF to be below 5 for all variables, with a few exceptions on the time dummies with maximum values below 8. The VIF results thus indicate that multicollinearity is not a particular concern in our analysis.

19 Our two subsamples of VC‐backed IPOs and TSX‐V graduations to the TSX behave quite similarly in terms of delisting. Among VC‐backed IPOs, there are two firms that delist within one year after the IPO, four firms that delist within two years after the IPO, eight firms that delist within three years after the IPO. Among the TSX‐V graduations to the TSX, there is one firm that delists within one year after the graduation, three firms that delist within two years after the graduation, and seven firms that delist within three years after the graduation.

20 We also replicate the one‐year analysis after dropping the three cases of firms that delist within the first year after the TSX listing, and find that our results are unaffected.

21 The estimates in the quantile regressions measure some conditional quantiles of our dependent variable (three‐year BHARs), given the predictor variables. In our table, we estimate coefficients for the first quartile (q = 0.25), for the median (q = 0.5), for the third quartile (q = 0.75), and for the threshold isolating the top 5% of performers (q = 0.95).

22 We thank an anonymous referee for pointing out this alternative interpretation.

Additional information

Notes on contributors

Michele Meoli

Michele Meoli is Assistant Professor in the Department of Management, Information and Production Engineering at the University of Bergamo.

J. Ari Pandes

J. Ari Pandes is Associate Professor in the Haskayne School of Business at the University of Calgary.

Michael J. Robinson

Michael J. Robinson is Associate Professor in the Haskayne School of Business at the University of Calgary.

Silvio Vismara

Silvio Vismara is Associate Professor in the Department of Management, Information and Production Engineering at the University of Bergamo.