4. RESEARCH RESULTS AND ANALYSIS
4.2 Empirical Results
4.2.3 Results for profitability after IPO and score of Risk Factors
Table 4-5 Estimation of Cox proportional hazards models Model 2(B)
Variables
Coeff. Hazard ratioVC 0.310 (0.42) 1.363
Overall Chi-square 34.73 (0.00) ***
The results reported include the coefficient of each control variables and the associated p-values in parenthesis for each model.*
significant at 10%, ** significant at 5%, *** significant at 1%. See Appendix 1. for variable definitions.
4.2.3 Results for profitability after IPO and score of Risk Factors
I now move on to an analysis of risk disclosure by scores. In a pure economy, disclosure quality monotonically reduces cost of capital, and firms with the highest disclosure ratings tend to also show the highest contemporaneous earnings performance (see Lang and Lundholm,1993). Therefore, I believe that the higher disclosure scores contribute the better earning performances in the future. However, Disclosure quality is very difficult to assess. As a result, in this analysis, I use a four-class index which attempts to assess the disclosure related to per risk factor presented among 20 items and the four subgroups.
In model 2(A), the result shows that , item8 (customer dependence) and item 19 (dependence on key employees)are significantly positive indicating that disclosing these two items increase the probability of post-IPO profitability. However, in model 3(A), a futher discussion, the result shows that the disclosure quality of item19 is not significant, and only item 8 is significantly positive at the 5% level indicating that disclosing this item well increases the probability of post-IPO profitability. It means investors care more about the revenue source of firms, customers. Besides, for
-
- -
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
48
continuous independent variables the hazard ratio represents the estimated percent change in the hazard of the event (attainment of profitability) for a one unit increase in the covariate of interest (controlling for other covariates) and is obtained by subtracting one from the hazard ratio and multiplying by 100. Therefore, I find that a one percent increase in the score of item8 results in a 77.9% increase in the probability of attaining profitability. This result is same with my expectation that investors care most about firms customers and want to know more details about this issue. Besides, the unsignificant result of score 19 might be due to the average score of item19 is 2.429.
Compared with other items, this is already a high score. This indicated that in my sample, many firms have already disclose detail information in item 19 (dependence on key employees), so the results don’t show significant relationship between score19 and probability of attaining profitability in the future.
‧
Table 4-6 Estimation of Cox proportional hazards models
The results reported include the coefficient of each control variables and the associated p-values in parenthesis for each model.*
significant at 10%, ** significant at 5%, *** significant at 1%. See Appendix 1. for variable definitions.
Variables Model 3(A)
Overall Chi-square 44.11 (0.00) ***
-
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
50
In addition to prior classification of of risk factors, I will discuss another classification, and how its score related with firm’s profitability after IPO.
In model 3(B), I find that only score of quantity risk factors has major impacts on the likelihood of profitability, althought in earlier results shows that whether disclosures of financial, quantity and uncertainty risk factors have signification effect on firm’s profitability after IPO. The coefficient of score of quantitative risk factors is significantly positive at the 1% level, indicating that disclosing this item increases the probability of post-IPO profitability, and about economic impact, one percent increase in the score of quantitative risk factors results in a 44.9% increase in the probability of attaining profitability.
I suggest that the release of a variety of quantative information about a firm (e.g., sales, net income, costs of operation, etc.) may be useful to investors, because investors know the firms more advanced, and can do more precise estimation about firm’s value.
When they more certain about firm’s value, they are willing to invest more money to firm, which increases the probability of post-IPO profitability. Besides, the unsignificant results about the score of uncertainty and financial risk factors may be for some reasons. Next I elaborate on the economic intuition for my result. In my research, average score of Uncertainty and financial are 3.762 and 4.476, respectively. However, the full scores of Uncertainty and financial are 20 and 12, respectively. It represents that firms disclose 18.81% information in firm’s future uncertainty, and 37.3%
information in firm’s financial issue, indicating that in general, firms are not used to disclose detail information in these two area. Therefore, we can not find a strong relationship between these two scores and firm’s probability of post-IPO profitability.
‧
Table 4-7 Estimation of Cox proportional hazards models Model 3(B)
Variables
Coeff. Hazard ratioVC 0.212 (0.58) 1.237
NUMEMP 0.001 (0.00) *** 1.001
FIRMAGE 0.110 (0.00) *** 1.116
LSIZE 0.005 (0.06) * 1.005
Financial Score 0.101 (0.52) 0.904
Quantity Score 0.371 (0.00) *** 1.449
Uncertainty Score 0.086 (0.21) 0.917
Downside Score 0.114 (0.13) 0.892
Overall Chi-square 31.82 (0.00) ***
The results reported include the coefficient of each control variables and the associated p-values in parenthesis for each model.*
significant at 10%, ** significant at 5%, *** significant at 1%. See Appendix 1. for variable definitions.
At last, in Figs.1-2 to 1-8, I graph the cumulative hazard functions by varying just the value of the covariate of interest while evaluating all the other covariates at their mean values. In my context, the cumulative hazard function tells us the cumulative risk of the profitability state being achieved over some period of time. Thus, the cumulative hazard function will always increase with time. For continuous variables, I plot the cumulative hazard function evaluating it first at the covariate's 25th percentile value and then at its 75th percentile value. For dummy variables, I evaluate the cumulative hazard function first at a value of zero for the covariate and then at a value of one for the covariate. Thus, this exercise allows us to visually see the impact of each variable on the cumulative hazard function holding the values of the remaining covariates at their mean values. I only plot these graphs for the variables that are significant (excluding significant control variables). These graphs indicate that about types of risk factors, cumulative hazard of attaining profitability increases with item8, item19 and quantitative risk factors, while it decreases with financial risk factors and incertainty risk factors. Futhermore, about scores of risk factors, cumulative hazard of attaining profitability increases with scores of item8 and quantitative risk factors. Thus, Figs. 1–
7 visually confirm my earlier reported findings.
- - -
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
52
Figure 1-2. Cumulative Hazard Function and Item
Figure 1-3. Cumulative Hazard Function and Item19
0.5 11.5 2
Cumulative Hazard Function
0 5 10 15 20
Time-to-Profitability (Quarters)
item8=1 item8=0
Cox proportional hazards regression
0.1.2.3.4.5
Cumulative Hazard Function
0 5 10 15 20
Time-to-Profitability (Quarters)
item19=1 item19=0
Cox proportional hazards regression
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
53
Figure 1-4. Cumulative Hazard Function and Financial Risk Factors
Figure 1-5. Cumulative Hazard Function and Quantitative Risk Factors
01234
Cumulative Hazard Function
0 5 10 15 20
Time-to-Profitability (Quarters) financial=1 financial=0
Cox proportional hazards regression
0
.02.04.06.08
Cumulative Hazard Function
0 5 10 15 20
Time-to-Profitability (Quarters) quantity=1 quantity=0
Cox proportional hazards regression
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
54
Figure 1-6. Cumulative Hazard Function and Uncertainty Risk Factors
Figure 1-7. Cumulative Hazard Function and Score of Item8
0.5 11.5
Cumulative Hazard Function
0 5 10 15 20
Time-to-Profitability (Quarters)
uncertainty=1 uncertainty=0
Cox proportional hazards regression
0.1.2.3
Cumulative Hazard Function
0 5 10 15 20
Time-to-Profitability (Quarters)
Score8 = 0 ( 25th percentile value = 75th percentile value)
Cox proportional hazards regression
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
55
Figure 1-8. Cumulative Hazard Function and Score of Quantitative Risk Factors
After looking all the graphs, they confirm my expectations about the relationship between risk factors in the prospectus and the probability of firms attaining post-IPO profitability. From fig 1-2 and fig 1-3, a biotech IPO with more information about main customers and key employees, will experience good performance after IPO, compared with firms not disclosing them. Firms disclosing item8 (customer dependence) are nearly six times as likely to attain post-IPO profitability compared to non-disclosure firms and firms disclosing item19 (dependence on key employees) are more than three times as likely to attain post-IPO profitability compared to non-disclosure firms. So there is a conclusion that disclosing more information about main customers than information about key employees has more effect on the probability of firms to attain post-IPO profitability. From figure 1-2 and figure 1-3, the gaps between firms disclosing them (item8=1, item19=1) and firms not disclosing them (item8=0, item19=0) are much more bigger for item 8, just like what I expect. Whether disclosing item8 or not has more influence on firms’future. The reasons might be that customers are the revenue sources of firms. They control firms’ profit margins, the key survival factor of firms.
Besides, about another classification of risk factors, firms disclosing financial risk factors are nearly half as less likely to attain post-IPO profitability compared to non-disclosure firms and firms disclosing uncertainty are more than half as less likely to
0.5 11.5
Cumulative Hazard Function
0 5 10 15 20
Time-to-Profitability (Quarters)
quantityscore=6 quantityscore=8
Cox proportional hazards regression
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
56
attain post-IPO profitability compared to non-disclosure firms. However, firms disclosing quantity risk factors are close to four times as likely to attain post-IPO profitability compared to non-disclosure firms. We can see that from figure 1-4 to figure1-6. The technological uncertainty about products in an development stage has higher litigation exposure (because of the uncertainty of success), and financial risk factors convey a signal to investor that firms were short of money to operate. Due to this, investors are relunctant to give their money to firms. Therefore, firms provide such information may deter them from raising money, leading to poor performance in the future. So there are negative relations exist in financial and uncertainty risk factors and probability of profitability after IPO. However, quantitative disclosures are valuable to investors, because it can let investors more certain about firm’s value and more willing to provide funds to firms. This contributes to the good performance of firms in the future, especially in the biotech industry, a competitive industry that need large amount of money to survive in it. So there is positive relation exists in quantitative risk factor and probability of profitability after IPO.
Furthermore, how firms disclose the items is also important. From figure 1-7 and 1-8, more detail descriptions in quantitative risk factors and customers’ information contributed to better performance in the future. A one percent increase in the score of item8 (customer dependence) results in a 77.9% increase in the probability of attaining profitability because investors care most about firms customers, revenue resources, and want to know details about this issue. They want to know whether firms have stable revenues and potential growth probabilities. Besides, one percent increase in the score of quantitative risk factors results in a 44.9% increase in the probability of attaining profitability. Quantative informations about a firm (e.g., sales, net income, costs of operation, etc.) are useful to investors, because investors are more certain about the operating situations of firms. Thus, they are willing to invest more money to firm, which increases the probability of post-IPO profitability. In a conclusion, not only whether firms disclose risk factors or not, but also how they disclose them are important. The risk factors in the prospectus are related to firms’ probability of profitability after IPO as expected.
‧
The development path of many emerging industries tend to be similar because they are characterized by high firm founding rates, rapid growth rates, substantial investments in R&D and capital expenditures, potential for product/process breakthroughs, investor exuberance, huge demand for capital, large number of firms going public while relatively young, and a struggle for survival during the post-IPO phase as profitability and growth targets remain elusive and shifts in investor sentiment substantially raise financing constraints. These industries are disk drives and biotechnology, and they attempt to go public at a relatively early stage in their development when there was still considerable uncertainty and information asymmetry regarding their future prospects. Because managers and insiders are better informed regarding a firm's prospects compared to outside investors. Therefore, investors aren’t willing to give out money to finance firms. However, these industries strongly need money to support their operating, so they disclose some valuable information in the prospectus to lower this uncertainty. Disclosures are market signals which are able to resolve information asymmetry between firms and investors. Consequently, managers and other insiders are in a position to credibly signal favorable information to outside investors regarding the likely future performance of IPO firms by disclosing more information to investors. Thus, they can raise fund from investors and lower the cost of capital, because investors can correctly value the firms. Furthermore, low costs of capital contribute to higher probability of profitability in the future.
My paper focuses on identifying risk factors disclosed in prospectus and their influences on the probability of biotech firms attaining post-IPO profitability. Because risk factors show us the potential operating risks of firms in the future, I think that they are more valuable to discuss with. Risks have close association with the probability of firms attaining post-IPO profitability. About research method of risk factors, I don’t focus on one way to classify. Instead, I use two kinds of classifications. One classification is that there are twenty types of risk factors based on prior studies. The other classification is that I combine some of them to form four subgroups. They are financial risk factors, quantitative risk factors, uncertainty risk factors, and downsides risk factors. I focus on two issues of them. First, I want to know whether disclose them or not affect the probability of firms attaining post-IPO profitability. Second, I also discuss how detail they disclose affect firms’ future earning probability.
After using survival analysis to run the data, my empirical results are that a biotech IPO with more information or some specific information of risk factors, like disclosures of main customers and key employees, will experience good performance after IPO.
The reasons might be that customers are the revenue sources of firms. They control