There is a strongly positive relationship between private equity and venture capital investment and intellectual property filings, with increased investment associated with increased likelihood of both patenting and trade marking activities.
Companies with a history of filing patents and trade marks are substantially more likely to continue these activities post-investment. Private equity and venture capital recognises the hallmarks of innovation and helps companies to become more competitive by enhancing their intellectual property portfolios.
This section presents the results of two logistic regression models that evaluate the likelihood of patenting and trade marking activities following private equity and venture capital investments. The dependent variables are binary indicators representing whether a company has a patent application or a trade mark application post-investment. The primary independent variables are the logarithm of the amount invested by PE firms (in thousands of euros) and the binary indicators for prior patenting and trade marking activities. Table 2 presents the results of the models.
Table 2
Summary of Propensity Models
Dependent variable: | ||
Subsequent patent activity (dummy) | Subsequent trade mark activity (dummy) | |
Amounts invested (logged) | 0.102*** | 0.122*** |
(0.007) | (0.005) | |
Previous patent activity (dummy) | 2.233*** | |
(0.025) | ||
Previous trade mark activity (dummy) | 1.723*** | |
(0.017) | ||
Controls: company sector and country | ||
| ||
Observations | 115,086 | 115,086 |
Log Likelihood | -25,761.790 | -46,276.890 |
Akaike Inf. Crit. | 51,615.590 | 92,645.780 |
Note: *p**p***p<0.01
To interpret the coefficients, we transform log-odds into odds ratios, which represent the probability of the event occurring divided by the probability of the event not occurring. The odds ratio is obtained by exponentiating the coefficient (eβ). Specifically:
- If the odds ratio is greater than 1, the odds of the outcome occurring increase as the predictor increases.
- If the odds ratio is less than 1, the odds of the outcome occurring decrease as the predictor increases.
- If the odds ratio is equal to 1, the predictor has no effect on the odds of the outcome occurring.
For patenting, the coefficient of amounts invested is 0.102, statistically significant at the 1% level, indicating that a 10% increase in the amount invested is associated with approximately a 1.02% increase in the odds of having a patent application post-investment.
The coefficient for prior patenting activity is 2.233, statistically significant at the 1% level. Exponentiating this coefficient, e2.233≈9.33, shows that companies with prior patenting activity have about 9.33 times higher odds to have a patent application post-investment compared to those without prior patenting activity.
For trade marking, the coefficient of amounts invested is 0.122, statistically significant at the 1% level, suggesting that a 10% increase in the amount invested is associated with approximately a 1.22% increase in the odds of having a trade mark application post-investment. The coefficient for prior trade marking activity is 1.723, statistically significant at the 1% level. Exponentiating this coefficient, e1.723≈5.60, indicates that companies with prior trade marking activity have about 5.60 times higher odds to have a trade mark application post-investment compared to those without prior trade marking activity.
The logit regression results demonstrate a strong positive relationship between the amounts invested by private equity and venture capital firms and the subsequent likelihood of patenting and trade marking activities by their portfolio companies. Specifically, increased investment is associated with higher odds of both patenting and trade marking activities. Moreover, companies with a history of IPR activity are substantially more likely to continue these activities post-investment.