The family holds more governance positions when it owns large stakes in small, profitable, low-risk firms. This result suggests that the family trades off expected costs and benefits by conditioning participation intensity on observable firm characteristics. We find that the positive effect of performance on participation is twice as strong as the positive effect of participation on performance.
Using ordered logit estimation, we find that the family participates more intensively the more equity it owns in the firm, the smaller the firm, the more profitable the firm, and the less risky the firm.
Examples of ordered logistic regression: A marketing research firm wants to investigate what factors influence the size of soda (small, medium, large or extra large) that people order at a fast-food chain. These factors may include what type of sandwich is ordered (burger or chicken), whether or not fries are also ordered, and age of the consumer. While the outcome variable, size of soda, is obviously ordered, the difference between the various sizes is not consistent. The difference between small and medium is 10 ounces, between medium and large 8, and between large and extra large 12.
We formalize this setting by an ordered logit model, where the dependent variable PARTICIPATIONit can take on four different values for firm i at time t. A higher value reflects higher intensity.