Private captive fund providers and the likelihood of achieving successful venture capital exits

ABSTRACT

We analyze the impact of the relationship between private captive venture capital firms (VCFs) and their parent investors on the likelihood of achieving successful exits in investee firms. We argue that the differences in strategic goals and the way venture managers are appointed and incentivized may affect the exit way achieved by different types of captive VCFs. In particular, we focus on private captive VCFs solely funded by a corporation or a financial institution, as well as on semi-captive VCFs jointly funded by several entities. We find that captive VCFs backed by corporations show a higher likelihood of exiting their portfolio firms more successfully than those backed by financial institutions and semi-captive VCFs. VCFs backed by corporations seem to have strategic goals that are compatible with the maximization of portfolio return, and their venture managers may contribute with valuable specific industry knowledge. Conversely, the managers of VCFs wholly owned by financial institutions rarely contribute with industry and technology knowledge and are even more affected by the lack of adequate compensation packages.

KEYWORDS:

• Venture capital
• captive investor
• agency theory
• parent investor
• managers
• exits

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. SPAINCAP (https://spaincap.org) is the new acronym of the Spanish Private Equity and Venture Capital Association. Webcapitalriesgo (https://www.webcapitalriesgo.com) is the service provider that conducts the annual surveys and updates the activity dataset about Spain on behalf of SPAINCAP and Invest Europe since 2002.

2. We also include them in the additional analyses carried out as further robustness checks.

3. In the US, the usual classification separates full acquisitions from secondary sales of the stakes held by VCFs to a third party (D. J. Cumming and MacIntosh 2003b), whereas in Europe trade sale is a category that includes both (e.g., see Invest Europe 2023). In addition, Invest Europe introduced a new category to compute sales to other VCFs. Hence, the number of acquisitions and secondary sales, following the US methodology, matches the sum of trade sales and sales to other VCFs in Europe.

4. Their data are based on a dataset before the separation of sales to other VCFs from trade sales.

5. The classification as a high-technology firm was taken from the original data provided by SPAINCAP, as this is a question requested and audited in the data-gathering process of the annual survey.

6. Other authors who analyze VC exits using a binary variable (e.g., Anderson, Chi, and Wang 2017; D. J. Cumming and MacIntosh 2003a), resort to logit models. Probit and logit models provide similar results, so the choice depends on the assumption regarding the probability distribution of the dependent variable, which is normal cumulative in probit and logistic in logit.

7. The estimation was carried out using linear models due to the impossibility of reaching convergence with IV Probit, caused by the existence of numerous fixed effects (Brander, Du, and Hellmann 2015).

8. The estimation of model 1A would allow us to indirectly test hypothesis 2 (CorpVCF versus SemicapVCF) by comparing the coefficients of the variables CorpVCF and SemicapVCF. However, we estimate model 1B, which is a variant of model 1A with the reference variable changed, to make the results more visible to readers. Nevertheless, the test of linear combination between both coefficients shows a positive significant difference (p-value <0.1) that coincides with the results of model 1B for CorpVCF.