On the relationship between IPO volume and underpricing. Identification of hot IPO markets using a bivariate regime switching model: Characterization of IPOs in hot markets

Chapter 1 of this dissertation investigates the bivariate time series behavior of industry-level quarterly Initial Public Offerings (IPO) volume and quarterly average IPO underpricing in the context of a bivariate Markov regime-switching model. The specification of our model allows us to objectively identify and date four regimes: Low Volume-Low Underpricing, High Volume-Low Underpricing, Low Volume-High Underpricing and High Volume-High Underpricing. We hypothesized that the regime High Volume-Low Underpricing is explained by a reduction in the level of asymmetry of information whereas the regime High Volume-High Underpricing is explained by an increase in the level of investor sentiment. We find that on the first half of the nineties all analyzed industries (with the exception of Depository Institutions) were in the regime High Volume-Low Underpricing whereas during the internet bubble period they were all in the regime High Volume-High Underpricing. Depository Institutions was in the regime High Volume-High Underpricing during the first half of the eighties a period that coincides with the deregulation of the banking sector.;Chapter 2 of this dissertation provides a characterization of IPOs in the two regimes associated to "hot markets" (High Volume-Low Underpricing and High Volume-High Underpricing). We concentrated our analysis in a group of variables that have been related in the IPO literature to the level of asymmetry of information surrounding an offering. These variables are: age of the issuing firm at the time of the IPO, size of the offering, offering price, price update and the presence of venture capital prior to the IPO. We find that some of the characteristics of issuing firms differ dramatically depending on the industry and on the regime. Also, we conduct a regression analysis to assess their explanatory power on the level of underpricing depending on the regimes and the industry. We find that the statistical significance and sign of the coefficients of the variables varies considerably depending on the industry and on the regime.