University of Amsterdam
Amsterdam Business School MSc Business Economics –Finance Track
Master Thesis
The effect of lowering IPO costs and legal requirements on firm
performance and job creation
Lynn Eileen Blokstra July 2016
Statement of Originality
This document is written by Lynn Blokstra who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Abstract
To stimulate economic growth and job creation the U.S. Securities and Exchange Commission (SEC) allows small firms to do their initial public offering (IPO) under reduced costs and loosen legal requirements. This thesis examines whether a low-cost IPO with loosen legal requirements can indeed spur economic growth and job creation by exploiting two settings: The Small Reporting Company Regulatory Relief and Simplification #33-8876 (SRC) rule and the Jumpstart Our Business Start-ups (JOBS) Act. Both deregulations were implemented using clear revenue cut-offs that allow me to identify similar counterfactual cases. In comparison to IPOs that are not subject to reduced costs and loosen legal requirements, the results show that the SRC rule and the JOBS Act have increased firm performance and employment at least until two years after IPO. Increases in market capitalization, assets, profitability and number of employees are found significant. Additionally, I find significant decreases in leverage ratio and increases in stock liquidity. The effect is stronger in the SRC rule and in which targeted firms were smaller. These findings provide insight in the success of an early IPO and the impact of deregulations on the economy.
Table of contents
1 Introduction ... 5
2 Literature Review and Institutional background ... 7
2.1 Going public and firms performance ... 7
2.2 Going public and job creation ... 8
2.3 The cost of going public and the effect of SOX ... 9
2.4 The SRC rule ... 10
2.5 The JOBS Act ... 10
2.6 Institutional Background ... 12
3 Data ... 12
3.1 Data Sources and Sample Description ... 12
3.2 Variable Construction... 14 3.3 Descriptive Statistics ... 15 4 Identification strategy ... 16 4.1 Empirical Method ... 17 5 Results ... 18 5.1 SRC rule ... 19
5.1.1 SRC rule: Real outcomes ... 19
5.1.2 SRC rule: Financial Outcomes ... 21
5.2 JOBS Act ... 22
5.2.1 JOBS Act: Real Outcomes ... 22
5.2.2 JOBS Act: Financial Outcomes ... 23
5.3 Discussion ... 23 6 Conclusion ... 25 References ... 27 Tables... 29 Figures ... 51 Appendix ... 56
1 Introduction
There are several reasons why going public is beneficial for firms (Cumming, 2012). However, it comes with a cost (Ritter, 1987). In theory, the initial public offering (IPO) costs work as a screening mechanism that separates ‘lemons’ from ‘cherries’. To bring the benefits of an IPO to small firms, the U.S. Securities and Exchange Commission (SEC) encourages an early IPO by providing reduced costs and loosen legal requirements for small issuers. In this way they want to stimulate economic growth and job creation (SEC, 2007 and U.S. Congress, 2012). The question rises whether an early IPO with reduced costs and loosen legal requirements could achieve their intended goals of economic growth and job creation. The aim of this thesis is to investigate whether lowering the costs and legal requirements of an IPO for small firms results in better firm performance and job creation in the medium to long term. On the one hand, reducing costs and loosing legal requirements may result in weaker companies going public. On the other hand, the benefits of the IPO itself may result in a positive effect on firm performance and job creation.
To test this hypothesis, I examine the effect of reducing costs and loosening legal requirements on firm’s real and financial outcomes. I exploit two settings, the first setting is the Small Reporting Company Regulatory Relief and Simplification #33-8876 (SRC) rule and the second setting is the Jumpstart Our Business Start-ups Act (JOBS) Act. The SRC rule became active on February 4th, 2008 and lowers costs and legal requirements for small IPOs by reducing disclosure requirements for small reporting companies (SRCs), which are firms with less than $50 million in annual revenue or $75 million in public equity floats (SEC, 2007). The JOBS Act was signed into law by Obama on April 5th, 2012. Title I of the act lowers costs and legal requirements by de-burdening and de-risking the IPO process for emerging growth companies (EGCs), which are firms with less than $1 billion in annual revenue (U.S. Congress, 2012).
The SRC rule and JOBS Act provide unique settings in which counterfactuals can be clearly identified. Both settings allow me to compare similar firms around the cut-off for SRC or EGC status. I make use of a diff-in-diff kink regression. First I compare the relationship between pre-IPO revenue and post-IPO performance for two groups of small firms; small firms going public after deregulations and small firms going public before deregulations. This model disentangles the problem that most firms are affected by the JOBS Act, which results in a limited control group of non-EGCs. The problem with the first model is that changes in the relation between pre-IPO revenue and post-IPO performance may happen for other reasons. Therefore, secondly I compare IPOs just above and below the cut-off. These firms are supposedly very similar, but the one below the cut-off is subject to lower costs and loosen legal requirements. If deregulation is effective, then the value of small firms below the cut-off must be less responsive to their size, because this indicates higher values of firm performance and job creation for small issuers than without deregulation.
Mikkelson et al. (1997), Jain and Kini (1999), Gao et al. (2013) and Davidoff Solomon and Rose (2014) already performed research on early IPO performance and find a drop in profitability
after IPO. The main reason is thought to be the size of the firm; small firms have a higher risk of failure than their larger counterparts. They also find poor lifecycle performance of small IPOs compared to large IPOs. These studies suggest that deregulations for small issuers might bring firms to the public markets that are unsuitable. On the other hand, the Sarbanes-Oxley Act (SOX) in 2002 especially increases costs on small public firms (Kamar et al.,2007), which could be an explanation of poor small IPO performance. Thereby, it is difficult to compare small IPOs and large IPOs, as these firms are very different. My study on early IPO performance makes use of a counterfactual analysis. It compares similar small IPOs subject to two different costs and legal requirements, which is more consistent. It contributes to the existing literature on early IPO performance and the relation between job creation and IPOs. Additionally, it contributes to the existing literature on the question whether the SRC rule and JOBS Act have the desired result for policy makers.
My findings suggest that a less costly IPO with loosen legal requirements has a positive effect on firm performance and job creation until at least two years after IPO. The effect stems probably from an increase in productivity and is stronger in the SRC rule and among small-targeted firms. The estimations show a significant positive effect on market capitalization, assets, profitability and the number of employees. In comparison to IPOs that are not subject to reduced costs and loosen legal requirements, market capitalization at least doubles for firms with at most $150 thousand in revenue prior to IPO. Two and three years after IPO, more firms are significantly affected. For job creation, one year after IPO the number of employees at least doubles for firms with at most $107 thousand in revenue prior to IPO. Two and three years after IPO, the same firms are affected, but the impact is even stronger. Furthermore, reducing costs and loosing legal requirement and in this way increasing access to capital markets helps to significantly reduce leverage ratios and increase stock liquidity.
Although the JOBS Act has only been in place for four years, which limits the sample, the examination gives a good indication of the effectiveness of deregulation. The results for the JOBS Act setting could be influenced by the SRC rule, as some firms already have reduced costs and loosen legal requirements. This influence is not present for the SRC rule setting, because the former Regulation S-B is eliminated by the SRC rule.
The results also give an idea about firms that did not do an IPO due to the costs and legal requirements, because the control group is selective. The control group does not include firms that decided not to go public. The results suggest that high performance firms are left out due to higher costs and legal requirements. Thereby, according to Dambra et al. (2015)1, loosening legal requirements is more likely to be the driver of the results. Reduced costs on the other hand do not seem to be the driver, as according to the existing literature deregulations do not reduce direct costs
and increase indirect costs (Chaplinsky et al., 2015)2. Deregulations give access to the benefits of an IPO to small firms, instead of encouraging weaker companies to go public.
My results show that the SRC rule and the JOBS Act are probably successful in achieving their intended goals of economic growth and job creation. However, policy makers should reflect all available information on deregulations.
The remainder of the thesis is structured as follows. Section 2 gives an overview of the existing literature and the institutional background. Section 3 describes the data. Section 4 details my identification strategy and empirical method. Section 5 presents the main empirical findings and discusses the results. Section 6 concludes the thesis.
2 Literature Review and Institutional background
This section presents the general literature related to IPO performance and deregulations. Particular emphasis is laid on small IPOs. Section 2.1 sets out the main literature on IPO performance. Section 2.2 describes the main literature of job creation in relation the IPOs. Section 2.3 gives an overview of the costs of going public and the effect of SOX. Section 2.4 presents relevant research on the SRC rule. Section 2.5 presents relevant research on the JOBS Act. Section 2.6 gives an overview of the institutional background.
2.1 Going public and firms performance
An initial public offering (IPO) is the first offering of stock by a private firm to the public. There are various reasons to go public, which could differently impact firm performance. One of the reasons of an IPO, which is widely stated in the literature, is to finance investments (Rydqvist and Högholm, 1995 and Chemmanur et al., 2009). On the other hand, Rydqvist and Högholm (1995) and Pagano et al. (1998) find evidence that suggests that firms go public to rebalance their capital structure after a period of high growth and investments. A third explanation, as discussed by Pagano et al. (1998) and Brau et al. (2003) could be a tool to ‘cash-out’ and create liquidity for insiders to diversify their portfolio. An additional motive of going public is to create a currency of shares to acquire other firms or to be able to get acquired (Zingales, 1995 and Brau et al., 2003). The motive of going public could also depend on the size and age of the firm (Pagano et al., 1998). Chemmanur et al. (2009) show that lower information asymmetry for older and larger firms increase the probability of an IPO.
Going public leads to the separation of ownership and control, known as the agency problem. According to Jensen and Meckling (1976), the separation of ownership and control, leads to fewer incentives for the management. This results in lower firm performance, compared to the period before the IPO. Jain and Kini (1994) support this evidence. They find that firms going public experience a decline in operating performance and show that higher retention of ownership by the original owners has a positive impact on operating performance relative to other issuers.
Mikkelson et al. (1997) tested the relation between ownership and operating performance of companies that go public and also find a decline in operating performance after going public. However, they suggest that this decline is unrelated to the ownership structure. They find that size and age of the firms play an important role. In fact, they find evidence that small and young companies especially underperform industry-matched firms, whereas the performance of larger firms is comparable to the performance of industry-matched firms after going public. In the long run (ten years), they find some improvements in the performance of small and young firms.
Jain and Kini (1999) conducted research on the life cycle of IPOs. They investigated three possible scenarios after IPO. It can continue as an independent firm, fail or get acquired. According to their research, the size of the firm is related to the risk of failure. Firms that go public too early in their life cycle are more likely to fail or to get acquired. Larger firms experience an increased probability of survival. Davidoff Solomon and Rose (2014) performed similar research, but focus on small IPOs. They find that small IPOs voluntarily or involuntarily delist, while their larger counterparts are more likely to get acquired. Moreover, small firms that remain listed fail to grow.
The existing literature suggests that it might be unsuitable for small firms to go public. Therefore, it is questionable whether encouraging small firms to go public with reduced costs and loosen legal requirements have the desired impact. By investigating this issue, this thesis contributes to the existing literature on IPO performance.
2.2 Going public and job creation
Among policymakers, slow employment growth in the period after the financial crisis is an important concern. Thereby, the drop in IPO volume is also an issue, since the volume is positively related to job creation. The decline in IPO volume has been most noticeable for small companies, and to circumvent this issue the JOBS Act was conceived (Borisov et al., 2015).
The capital raised at an IPO can be used to hire new employees. Kenney et al. (2012) examined job creation of IPOs for U.S. companies that went public between June 1996 and December 2010. They find that the average firm creates 822 jobs since its IPO. After ten years of IPO, employment increases by 60%, this is 4.8% annual growth, using the compounded rate. Ritter (2014) uses these numbers to calculate the number of jobs lost due to the drop in IPO volume. He finds that 1.87 million jobs are lost due to the low IPO volume. Weild and Kim (2009) performed similar research. They find that there are 22.7 million not created due to the low IPO volume. The difference in statistics between Weild and Kim (2009) and Ritter (2014) is caused by different assumptions. Similarly, Borisov et al. (2015) show that firms almost double the number of employees in three years after IPO. They also show that this is mainly caused by better access to public capital.
The existing literature on the relation between going public and job creation is conclusive. There is an important positive relation between public capital markets and job creation. Going public results in job creation. However, the literature does not make a distinction between small IPOs and
large IPOs and does not examine the effect of deregulations for small IPOs on job creation, which is the contribution of this thesis.
2.3 The cost of going public and the effect of SOX
Ritter (1987) presents evidence on the cost of going public. The two main components are direct costs and indirect costs. Direct costs are fixed costs, which are mainly investment banking fees. The main indirect cost is underpricing, due to information asymmetry. It is a risk premium, which is a compensation for investors caused by uncertainty about the firm’s issuing value. IPOs are underpriced more the greater the ex ante uncertainty about an issuing firm’s value. According to Verrechia (2001) higher cost of capital is a result of greater underpricing, due to reduced disclosures.
The SOX in 2002 forced additional regulatory costs on public firms in the U.S.. The primary goal of SOX is to improve transparency in public companies by increasing disclosure and monitoring requirements. Whether the benefits of SOX outweigh the costs is inconclusive (Kaserer et al., 2011). Some argue that the impact of SOX is different on small firms than on large firms. Both the costs and the benefits of complying with SOX could be larger for small firms. The costs could be relatively higher, because some costs are fixed. However, the benefits of confidence and transparency could also be higher (Kamar et al., 2007).
According to the overview of different studies on the impact of SOX on the costs for public firms by Kamar et al. (2007), SOX increased public firms’ regulatory costs, both for small and large firms. The increase in costs is larger for small firms. Kaserer et al. (2011) examined the impact of SOX on the direct and indirect cost of going public. According to their results, the direct cost of going public increases, but the indirect costs, measured by underpricing, decreases due to reduced asymmetric information. The overview of evidence on the impact of SOX on firm value by Kamar et al. (2007) is mixed, however the studies are consistent in finding a negative impact on smaller firms.
Many have blamed that SOX regulation and the high cost also negatively impacts IPO volume. The IPO volume during 2001-2016 is far lower than the IPO volume in 1980-2000. Consistent with the increase of cost due to SOX, the decline in IPO volume has been most noticeable for small firms. The decline in IPO volume could have negative effects on Gross Domestic Product (GDP) growth and job creation. As already mentioned, the JOBS Act was signed into law in order to counter this issue (Gao et al., 2013).
Besides increased costs of going public forced by SOX, there could be other reasons for the drop in small firm IPO volume. Gao et al. (2013) give the ‘economies of scope’ theory as explanation. According to them, a change in the economy caused a decrease in profitability for small firms, either public or private. They argue that small firms could create greater operating performance by getting acquired instead of relying on organic growth. They also find that the probability of small issuers getting acquired after IPO has increased over time. Davidoff Solomon and Rose (2014) argue that it is caused by market conditions; there is low demand for investments in small IPOs due to high risk and
low return. They claim that small firms might be unsuitable to go public. On the other hand, they show that there is an increase in voluntary delisting and takeovers among small firms after SOX, which indicates that SOX might have a negative effect on small firms.
According to the literature overview, SOX increases costs for small public firms and negatively impacts firm value and small firm IPO volume. On the other hand, different theories argue that SOX is not the driver of poor small firm performance and the drop in small firm IPO volume. Therefore, it is questionable whether further deregulations are able to increase IPO volume and achieve their intended goals: economic growth and job creation. Research on the SRC rule and the JOBS Act is already performed, which will be discussed in the next section.
2.4 The SRC rule
Ritter (2014) did research on the impact of the reduced disclosure requirements for small issuers on IPO volume, by exploiting the SRC rule for firms with public equity float below $75 million. They find evidence that there is no change in IPO activity following this regulatory change.
Cheng et al. (2013) conducted research on the effects of reduced disclosure requirements for firms with a public equity float smaller than $75 million on the information asymmetry component of the costs of capital, which they proxy with market illiquidity. They make a distinction between voluntary and mandatory disclosure requirements. They find that firms that are qualified to reduce disclosures, but voluntarily disclose information, experience a decrease in market liquidity. This effect is larger for firms with higher information asymmetry. This study suggests that mandatory disclosure serves as a more credible signal and cannot be replaced by voluntary disclosure, as this results in higher illiquidity, which is costly.
Gao et al. (2009) show that there could be negative side effects when small companies are spared from disclosure requirements. They did research for firms with public equity float smaller than $75 million and find that firms have incentives to remain small, to stay below the threshold of $75 million public equity float. They show that these firms undertake less investments and report lower earnings than control firms.
The literature overview does not show a positive impact of reduced disclosure requirements under the SRC rule. However, no research exists on the overall effect of reduced disclosure requirements for small issuers under the SRC rule. This thesis fills this gap by investigating the impact on firm performance and job creation. The literature mainly focuses on firms with public equity float of less than $75 million, while this thesis focuses on IPOs with revenue smaller than $50 million prior to IPO.
2.5 The JOBS Act
Dambra et al. (2015) conducted important research on the JOBS Acts and IPO volume. After controlling for market conditions, they estimate that the JOBS Act results in a 25% increase of annual IPOs over the level before the JOBS Act, especially among small firms. Most of the increase in IPO
volume is concentrated among firms with high proprietary disclosure costs, such as biotechnological and pharmaceutical firms. These firms benefit most from reduced costs and loosen legal requirements. The result suggests that the JOBS Act has been effective in encouraging small firms to go public. They also find evidence that the de-risking provisions instead of the de-burdening provisions are the main driver of the results. This indicates that costs are not the main driver in the decision.
However, according to the research of Dambra et al. (2015), the long-term performance of these firms is unsure. They find that the average small firm going public since JOBS Act is lossmaking and thereby the auditors doubted for approximately one third of these firms whether they would be able to continue as a going concern. Therefore, this thesis contributes to the study of Dambra et al. (2015) by examining the medium to long-term outlook, as more time has passed since the implementation of the JOBS Act.
Zimmerman (2015) forms evidence on the corporate governance characteristics of companies that are the first with EGC status and whether these companies adopt the JOBS Act’s reduced disclosure requirements. According to the research of Zimmerman (2015), the smallest firms adopt the JOBS Act’s reduced disclosure requirements. This indicates that the targets of the act, smaller companies, react in favor of the exemptions. However, smaller reporting companies already have reduced disclosure requirements under the SRC rule.
Similar to Dambra et al. (2015), Zimmerman (2015) also shows that the costs are not the driver of taking reduced disclosure requirements. The estimations show a significant positive relation between corporate governance and the decision of foregoing reduced disclosure requirements. The results show that greater board independence and audit committee accounting expertise increases the probability of foregoing reduced disclosure requirements. Unintended consequences could be that less transparency could result in moral hazard and adverse selection.
The fact that costs do not seem to be the issue, which is shown by Dambra et al. (2015) and Zimmerman (2015) is supported by the evidence of Chaplinsky et al. (2015). They examine the effect of the JOBS Act on the direct costs (issuance, accounting, legal and underwriting fees) of going public, but do not find a reduction for EGCs. Thereby, critics of the JOBS Act notify that reducing disclosure requirement could result in greater information asymmetry for investors. To compensate for this uncertainty, it could result in a higher risk premium investors demand. Chaplinsky et al. (2015) show that the indirect costs of issuance, namely underpricing, is significantly higher for firms affected by the JOBS Act than other IPOs. Barth et al. (2014) show similar results. However, according to Chaplinsky et al. (2015), there should be other benefits that outweigh the costs, as most firms claim EGC status. This thesis examines other benefits and contributes to the existing literature on the JOBS Act, by providing insight in the overall effect on firm performance and job creation.
2.6 Institutional Background
Before the introduction of the SRC rule, Regulation S-B was adopted in U.S., in 1992. Firms with revenue and public equity float smaller than $25 million were affected by this rule. The goal of this act was to increase access to public capital markets for small issuers. Regulation S-B was one of the first rules that reduce costs for small issuers. However, after the SOX in 2002, disclosure costs have sustained to be debated. Therefore, the SEC came with SRC rule #33-8876. This rule eliminates Regulation S-B (Cheng et al., 2013 and SEC, 2007).
The SRC rule #33-8876 came into effect on February 4th, 2008. It provides reduced costs and loosen legal requirements for small IPOs by reducing disclosure requirements for SRCs, which are firms with less than $50 million in annual revenue or $75 million in public equity float. It allows choosing reduced disclosures on ten non-financial items. Firms are free to make use of any, all, or none of the reduced disclosures. The new rule is aimed to increase the number of small firms affected by the reduced disclosure requirements and thereby to further reduce costs (SEC, 2007). The reactions on the SRC rule are mixed. Small companies generally respond in favor of the reduced disclosure requirements grounded on the potential of reduced costs of going public. However, some investors and accounting firms were worried, as it may give incentives to only report optimistic information (Cheng et al., 2013). A summary of the reduced disclosure requirements is provided in Appendix A.
In addition to the SRC rule, the JOBS Act came into effect on April 5th,2012. Title I of the act encourages an early IPO, as it reduces costs and loosens legal requirements by burdening and de-risking the IPO process for small issuers, known as EGCs. EGCs are firms with annual revenue smaller than $ 1 billion, so the JOBS Act affects larger IPOs than the SRC rule. The main goal is economic growth and job creation (U.S. Congress, 2012). Similar to the SRC rule, when an issuer qualifies as EGC, it can choose to take advantage of any, all, or none of Title I’s provisions (Dambra et al., 2015). In contrast to the SRC rule, the JOBS Act is intended specifically at companies going public (Zimmerman, 2015). This thesis follows the same classification of provisions as Dambra et al. (2015): de-risking provisions and de-burdening provisions. A summary is provided in Appendix B.
3 Data
3.1 Data Sources and Sample Description
The data comes from COMPUSAT and represents U.S. IPOs of common stock between 2000 and 2016. For 2016, only the first four months are included. From this dataset, two samples are created. The first sample, the SRC rule sample, represents U.S. IPOs between 2000 and 2016. The date range is eight years before and eight years after the SRC rule came into force. The second sample, the JOBS Act sample, represents U.S. IPOs between 2003 and 2016. The sample starts at 2003, because at the beginning of 2003 issuers are affected by additional costs associated with SOX (Chaplinsky et al., 2015).
For both samples, offerings in the financial sector are excluded. Thereby, the sample is restricted to firms for which there is financial information available in the fiscal year prior to IPO. It is not required that these firms have post-IPO financial data on COMPUSTAT. Consequently, the number of firms in the sample varies for different time windows. The first sample consists of 1,669 public offerings between 2000 and 2016. The second sample consists of 1,536 public offerings between 2003 and 2016. It is possible that the data selection criteria may have caused a survivorship bias. The distribution of the IPOs over years for both samples is shown in Figure 1 panel A and B.
Figure 1 shows a substantial increase in IPO volume after 2003. The figure shows that 2004 is almost three times higher than 2003. It is the highest number of offerings since the dot com bubble. It also shows a substantial decrease after 2007. The number of IPOs during the global financial crisis (2008-2009) is near zero. Issue volumes began to improve in 2010 and go back to around pre-crisis level in 2013/2014.
Figure 1 panel A shows the first sample. The sample is divided between firms with revenue smaller and larger than $50 million in the fiscal year prior to IPO (SRCs and non-SRCs). The sample consists of 638 SRCs and 1,031 non-SRCs. The figure shows that the number of SRC IPOs lies under the number of non-SRC IPOs, except for 2000 and 2001. The SRC rule came into force during the financial crisis. The figure does not show a substantial increase of SRC IPOs after the rule came into force. After the financial crisis the number of IPOs is improving, however issues of SRCs are not substantial more compared to pre-crisis levels. In the sample for the period for which the SRC rule is in effect, 32% of all IPOs are SRCs.
Figure 1 panel B shows the second sample. This sample is divided between firms with revenue smaller and larger than $1 billion in the fiscal year prior to the IPO (EGCs and non-EGCs). The sample consists of 1,360 EGCs and 176 non-EGCs, showing that the threshold is sufficiently high that most firms succeed for the EGC status. The figure shows that after passage of the act, IPO volume increases. It seems that after the JOBS Act came into force, there is an increase in IPOs for EGCs. In 2015 and the beginning of 2016, IPO volume falls abruptly. In the sample period for which the act is in effect, 87% of all IPOs are EGCs.
The performance of IPOs is measured one, two and three years after IPO. Table 1 provides information for the number of firms for which it is possible to construct financial measures for different time windows. Both samples are divided into four groups; the period before and after the SRC rule or the JOBS Act came into force and thereby between firms that have revenue smaller and larger than $50 million for the SRC rule sample (SRCs and non-SRCs) or revenue smaller and larger than $1 billion for the JOBS Act sample (EGCs and non-EGCs) in the fiscal year prior to the IPO. In the period before the SRC rule came into force the number of firms is quite stable over time, for SRCs and non-SRCs. However, in the period after the SRC rule came into force, the number of firms in the sample is decreasing if time passes after IPO. This could be due to low IPO volume directly after the SRC rule was implemented. The JOBS Act sample shows similar patterns as the SRC rule sample.
The number of firms in the sample is especially small three years after IPO for firms that went public after the JOBS Act, both for EGCs and non-EGCs, probably because the JOBS Act is only in place for four years. The sample for non-EGCs is even smaller, caused by the high revenue cut-off.
3.2 Variable Construction
There are variables constructed for the analyses, which are provided in Table 2. To test the medium to long-term firm performance and job creation, this study constructs variables one, two and three years after the IPO. To measure firm performance, different dependent outcome variables are constructed. These dependent outcome variables are divided between real outcomes and financial outcomes. The main real outcome variable is MarketCap, measured by the log of common shares outstanding (COMPUSTAT csho) times fiscal year closing price (COMPUSTAT prcc_f). Other real outcome variables might be able to explain the impact on MarketCap. These outcome variables are: Assets, measured by the log of total assets (COMPUSTAT at); Investment, measured by capital expenditures (COMPUSTAT CapEx) over lagged total assets; and to measure efficiency Sales/Assets, measured by sales (COMPUSTAT sale) over lagged total assets.
Another real outcome variable, which is used by Jain and Kini (1994) to measure operating performance is operating cash flows (CF/Assets), measured by operating income (COMPUSTAT oibdp) minus capital expenditures over lagged total assets (COMPUSTAT at). This is a good measure for operating performance, because cash flows are a primary component of the valuation of a firm.
Employees is an additional real outcome variable, measured by the log of employees
(COMPUSTAT emp). This variable is important, as it is a measure for job creation.
Besides real outcome variables, three financial outcome variables are constructed. The market to book ratio (M/B), measured by the log ratio of market value of equity (COMPUSTAT csho times prcc_f) over book value of equity (COMPUSTAT ceqt); Leverage measured by total debt (COMPUSTAT dltt + dlc) over total assets; and Liquidity, measured by the log ratio of volume (COMPUSTAT cshtrm) over shares outstanding (COMPUSTAT cshom). Liquidity is used as dependent outcome variable, as according to several authors, for example Cooray (2010), market liquidity contributes to economic growth.
Besides dependent outcome variables, independent variables are also constructed. In both cases the running variable is Size, measured by the log of revenue in the fiscal year prior to IPO. For the SRC rule, the cut-off is $50 million revenue, while for the JOBS Act the cut-off is $1 billion in revenue. Besides the independent variable, two dummy variables are constructed. A dummy variable (D1) indicating one for IPOs after February 4th, 2008 for the SRC rule setting and after April 5th, 2012 for the JOBS Act setting. A dummy variable (D2) indicating one if the firm went public with revenue below the revenue cut-off in the fiscal year prior to IPO.
A set of control and other variables are also constructed. The analysis accounts for industry differences (Industry), based on two-digit SIC codes (COMPUSTAT sic) and the capital structure in
the fiscal year prior to IPO, Leverage-1, measured by total debt over total assets. To get an insight in the characteristics of firms that go public, the age of the firm (Age)3 is constructed, measured by IPO year (COMPUSTAT ipodate) minus founding year.
To remove the potential influence of extreme measures of performance, the data is winsorized by setting the top and/or bottom at maximum of 5 per cent, depending on the distribution of the values.
3.3 Descriptive Statistics
Summary statistics for the samples are presented in Table 3 for the SRC rule and Table 4 for the JOBS Act. Both tables have a similar set-up. Following the schedule of deregulations, the samples are divided into four groups. Firms that went public before and after the SRC rule or the JOBS Act came into effect. These different time samples are divided between firms that have revenue smaller and larger than $50 million for the SRC rule sample (SRCs and non-SRCs) or revenue smaller and larger than $1 billion for the JOBS Act sample (EGCs and non-EGCs) in the fiscal year prior to the IPO. Panel A represents the sample for IPOs before and after the SRC rule or the JOBS Act came into effect, with firm characteristics one year before IPO. Panel B represents IPOs before and after one of the deregulations, with real and financial outcome variables one year after IPO.
Table 3 panel A shows that SRCs are smaller (Size) and younger (Age) than non-SRCs, both before and after the SRC rule came into effect. Panel B shows that the mean of all performance measures, both real outcomes and financial outcomes are smaller for SRCs than for non-SRCs one year after IPO, both before and after the SRC rule came into effect. Overall, it indicates that non-SRCs perform better than SRCs.
Comparing SRCs before and after the rule was implemented; the table shows that SRCs that went public after the SRC rule are smaller (Size) and younger (Age) than SRCs that went public before the SRC rule. This indicates that the SRC rule encourages small and young firms to go public. One year after IPO, the real and financial outcome variables for SRCs that went public after the SRC rule are smaller than SRCs that went public before the SRC rule, except for MarketCap, Assets, and
Liquidity. Therefore, there is no clear pattern whether SRCs performs better after the SRC rule came
into effect.
The JOBS Act sample shows similar characteristics as the SRC rule sample. Table 4 panel A also shows that EGCs are smaller (Size) and younger (Age) than non-EGCs. Thereby, panel B shows that one year after IPO non-EGCs perform better than EGCs. However, the difference with the SRC rule sample is that Investment is larger for EGCs than for non-EGCs, both before and after the JOBS Act came into effect. M/B and Liquidity is also larger for EGCs than for non-EGCs after the JOBS Act came into effect.
3 Data on age of the IPOs is collected from Jay Ritter’s IPO data web site. This database gave information on the founding date of all IPOs between 1975 and 2015. Updated, March 2016.
Comparing EGCs before and after the JOBS Act came into effect also shows similarities with the SRC rule sample. This indicates that also the JOBS Act might encourage small and young firms to go public. Similar to the SRC rule, there is no clear pattern whether EGCs perform better one year after IPO after the JOBS Act came into effect. The difference with the SRC rule sample is that besides
MarketCap, Assets and Liquidity also Employees, M/B and Leverage are larger for EGCs that went
public after the JOBS Act than EGCs that went public before the JOBS Act.
4 Identification strategy
The estimation strategy exploits two settings to test the hypothesis. The first setting uses the SRC rule. The second setting uses the JOBS Act. Both deregulations make an early IPO more attractive, by providing reduced costs and loosen legal requirements for small issuers. As the JOBS Act has only been in effect for four years, there is limited data to measure the longer-term performance of an early IPO. This limitation is not present for the SRC rule setting. Thereby, for the JOBS Act the cut-off of $1 billion revenue is sufficiently high, that there are few firms that do not qualify for EGC status (Table 1 and Figure 1), which results in a limited sample for non-EGCs. The cut-off for the SRC rule is lower, $50 million revenue. Consequently, the former setting is more equally divided between SRCs and non-SRCs (Table 1 and Figure 1).
The introduction of the SRC rule and the JOBS Act provides a natural experiment to examine the performance of an early IPO with reduced costs and loosen legal requirements. Both deregulations provide unique settings in which counterfactuals can be clearly identified. It is an ideal setting to identify a causal effect of going public early with reduced costs and loosen legal requirements on firm performance and job creation. First, the SRC rule and the JOBS Act created an exogenous shock in the regulation. Second, it did not contain a prolonged jurisdictive and implementation process. The SRC rule was introduced in July 2007, in December 2007 the SEC released the final rule and two months later it became active (Cheng et al. 2013). For the JOBS Act, the jurisdictive period was only five moths. The act was introduced in congress in December 2011, and was signed into law five months later in April 2012 (Chaplinsky et al., 2015).
Firms that went public after February 4th, 2008 with revenue smaller than $50 million in the fiscal year prior to IPO are affected by the SRC rule. Firms that went public after April 5th, 2012 with revenue smaller than $1 billion in the fiscal year prior to IPO are affected by the JOBS Act. To estimate the effect of an early IPO with reduced costs and loosen legal requirements on performance I compare the relationship between pre-IPO revenue and post-IPO performance for small firms (SRCs and EGCs) going public after deregulations with small firms going public before deregulations. Additionally, I compare IPOs just above and below the revenue cut-off. These firms are very similar, but the one below the cut-off is affected by deregulations. This results in a combination of a difference-in-difference design and a regression discontinuity, namely a diff-in-diff kink regression.
Figure 2 represents the diff-in-diff kink regression for different outcome variables for the SRC rule setting one, two and three years after IPO. Panel A shows a real outcome variable, MarketCap as performance measure. Panel B shows a financial outcome variable, M/B as performance measure. As can be seen from the figures, there is a clear discontinuity between SRCs and non-SRCs after the SRC rule came into effect, this discontinuity is not present before the SRC rule came into effect.
Figure 3 represents the diff-in-diff kink regression for the JOBS Act setting, with similar set-up and outcomes variables as the SRC rule setting. The figures show a discontinuity for the
MarketCap and M/B between EGCs and non-EGCs after the JOBS Act, which is less present between
EGCs and non-EGCs before the JOBS Act came into effect. Comparing the SRC rule setting and the JOBS Act setting shows that the discontinuity is stronger in the SRC rule setting.
4.1 Empirical Method
Firm performance and job creation is analysed until three years after IPO. The study makes use of a diff-in-diff kink regression. A linear relation is specified between pre-IPO revenue and post-IPO performance for treated early IPOs with reduced costs and loosen legal requirements (SRCs or EGCs) caused by deregulations and control IPOs without reduced costs and loosen legal requirements (non-SRCs or non-EGCs). This results in two models, outlined in equation 1 and 2:
𝑦𝑖𝑡 = 𝛼 + 𝛽1𝑆𝑖𝑧𝑒𝑖𝑡+ 𝛽2(𝑆𝑖𝑧𝑒𝑖𝑡× 𝐷1)+𝑍𝑖𝑡+ 𝜀𝑖𝑡 𝑖𝑓 𝐷2 = 1 (1)
𝑦𝑖𝑡 = 𝛼 + 𝛽1𝑆𝑖𝑧𝑒𝑖𝑡+ 𝛽2(𝑆𝑖𝑧𝑒𝑖𝑡 × 𝐷1) + 𝛾1𝐷2 + 𝛽3(𝑆𝑖𝑧𝑒𝑖𝑡× 𝐷2) (2) +𝛾2(𝐷1× 𝐷2)+ 𝛽4(𝑆𝑖𝑧𝑒𝑖𝑡 × 𝐷1× 𝐷2) + 𝑍𝑖𝑡 + 𝜀𝑖𝑡
𝑦𝑖𝑡 is a performance measure one, two and three years after IPO (Year +1, +2, +3). 𝑆𝑖𝑧𝑒𝑖𝑡 represents the log of revenue – cut-off in the fiscal year prior to IPO. In the SRC rule setting, the cut-off is $50 million revenue, while in the JOBS Act setting the cut-off is $1 billion revenue. 𝐷1 is a dummy variable that is one for firms that went public after the SRC rule in the first setting and after the JOBS Act in the second setting and zero for other IPOs. 𝐷2 is a dummy variable that is one for firms that went public with revenue below the cut-off in the fiscal year prior to IPO and zero for other IPOs.
𝑍𝑖𝑡 are control variables. The regression controls for Industry differences, based on two-digit SIC code. There are two main reasons to control for industries, the first reason is that performance could be different across industries and this is a normal approach for these performance measures (Jain and Kini, 1994). Additionally, according to the research of Dambra et al. (2015), firms with high proprietary disclosure costs, such as biotechnology and pharmaceutical firms, increase IPO activity the most after the JOBS Act, as these firms are most affected. The regression also controls for the capital structure prior to IPO, Leverage-1 and year fixed effects.
revenue below the cut-off in the fiscal year prior to IPO (SRCs or EGCs). The parameter of interest is 𝛽2, which assesses the impact of deregulations (D1=1) on the relation between 𝑆𝑖𝑧𝑒𝑖𝑡 and 𝑦𝑖𝑡, in comparison to IPOs that are not affected as they go public before deregulations (D1=0). The response is related to the size of the firm prior to IPO.
The problem with Model (1) is that changes in the relationship between 𝑆𝑖𝑧𝑒𝑖𝑡 and 𝑦𝑖𝑡 may happen for other reasons. Accordingly, Model (2) does not restrict the sample and uses the revenue cut-off to compare firms around the cut-off. These firms are supposedly very similar, but firms going public with revenue below the cut-off are affected by deregulations, while firms going public with revenue above the cut-off are not affected by deregulations.
There are two parameters of interest in Model (2), 𝛾2 and 𝛽4. Model (2) shows parameter 𝛾2, which Model (1) does not show. 𝛾2 and 𝛽4 assess the impact of deregulations on 𝑦𝑖𝑡 in comparison to IPOs that are not affected by deregulations. Firms that are affected are firms that go public after deregulations (D1=1) and are small firms, SRCs or EGCs (D2=1). Firms that are not affected are firms that either go public before deregulations (D1=0) or are large firms going public, SRCs or non-EGCs (D2=0). 𝛾2 assesses the shift in the relation caused by deregulations independent on the size of the firm prior to IPO. 𝛽4 assesses the same as the parameter 𝛽2 in Model (1), it assesses the shift in the relation caused by deregulations dependent on the size of the firm prior to IPO. If deregulation is effective, the value of small firms below the cut-off must be less responsive to their size. The combination of 𝛾2 and 𝛽4 assesses the total effect of deregulations on firm performance and job creation.
The combined effect for Model (2) on a performance measure (𝑦𝑖𝑡) for Model (2) of going public early with reduced costs and loosen legal requirements caused by deregulations is assessed using: 𝛾2+ 𝛽4 × 𝑆𝑖𝑧𝑒𝑖𝑡, with changing values of 𝑆𝑖𝑧𝑒𝑖𝑡. The significant combined effect depends on the size of the firm prior to IPO, 𝑆𝑖𝑧𝑒𝑖𝑡.
Model (1) is more efficient, while Model (2) is unbiased. In other words, Model (1) gives a lower variance of the estimator of interest, however it might suffer from omitted variable bias. On the other hand, Model (2) does not suffer from omitted variable bias, but the variance of the estimator of interest increases. To investigate whether Model (1) is still reliable, a Chi-Squared test is performed to approve that Model (1) and Model (2) predict the same. This test measures whether the estimations for the parameters of interest, 𝛽2 for 𝑆𝑖𝑧𝑒𝑖𝑡× 𝐷1 in Model (1) and 𝛽4 for 𝑆𝑖𝑧𝑒𝑖𝑡 × 𝐷1× 𝐷2 in Model (2), are not statistically different.
5 Results
This section is divided into two parts, one for the first setting: The SRC rule and the second for the other setting: The JOBS Act. The goal is to find the impact of an early IPO with reduced costs and loosen legal requirements on firm performance and job creation. Firm performance is measured with
different performance measures. The real outcomes are presented in section 5.1.1 for the SRC rule setting and in 5.2.1 for the JOBS Act setting: 1. MarketCap, which could be explained by 1.1 Assets 1.2 Investment 1.3 Sales/Assets 1.4 CF/Assets. Also the impact on Employees is measured. The financial outcomes are presented in section 5.1.2 for the SRC rule setting and 5.2.2 for the JOBS Act setting: 2. M/B 2.1 Leverage 2.2 Liquidity. Section 5.3 discusses the results of both settings. The preceding analysis finds that an early IPO with reduced costs and loosen legal requirements has a positive effect on firm performance and job creation at least until two years after IPO.
5.1 SRC rule
In Table 5 till 5.8 the regression outcomes of the SRC rule setting for the real and financial outcome variables of Model (1) (column 1,2 and 3) and Model (2) (column 4,5 and 6) one, two and three years after IPO are presented. All regressions control for leverage, industry and year fixed effects. The Chi-Squared statistics are presented on the bottom of the tables, which tests equality for the coefficients of the variables of interest in Model (1), Size x D1 and Model (2), Size x D1 x D2.
In addition to the tables, Figure 4 presents a change in relation caused by an early IPO with reduced costs and loosen legal requirements under the SRC rule between the size of the firm in the fiscal year prior to IPO and a performance measure. This is presented one, two and three years after IPO. It shows the confidence interval of the combined effect for Model (2) of a shift in the relation dependent and independent on the size of the firm prior to IPO.
5.1.1 SRC rule: Real outcomes
Table 5 presents MarketCap as performance measure. The results show that the SRC rule has a positive effect on MarketCap one, two and three years after IPO. The effect depends on the size of the firm prior to IPO; small SRCs are more affected. This is shown in the table. One, two and three years after IPO, the results show a negative and statistically significant coefficient at least at a 10% level for the variables of interest for Model (1), Size x D1 and at least at a 5% level for Model (2), Size x D1 x D2. The Chi-squared test shows that these coefficients of Model (1) and Model (2) are not statistically different. In addition to Model (1), Model (2) is able to present the impact of the SRC rule on
MarketCap unrelated to the size of the firm prior to IPO with D1 x D2. The coefficient is insignificant,
indicating that not everyone is equally affected.
The confidence interval for the combined effect on MarketCap for Model (2) is shown in Figure 4 panel A. In comparison to IPOs not affected by the SRC rule, one year after IPO MarketCap at least doubles for firms with at most $150 thousand in revenue prior to IPO. For two and three years after IPO, more firms are significantly affected. Two years after IPO, MarketCap at least doubles for firms with at most $900 thousand in revenue prior to IPO. After three years, MarketCap increases by at least 80 per cent for firms with at most $8.7 million revenue prior to IPO. The findings provide evidence that going public early with reduced costs and loosen legal requirements increases
Assets and profitability (CF/Assets) explain the impact on MarketCap. The findings provide
evidence that an early IPO with reduced costs and loosen legal requirements positively influences
Assets and CF/Assets one and two years after IPO. The effect depends on the size of the firm prior to
IPO; small SRCs are more affected. There is no effect that equally impacts every firm.
The results for Assets as dependent outcome variable are shown in Table 5.1. The table shows that the coefficient for Size x D1 for Model (1) and for Size x D1 x D2 for Model (2) are negative and statistically significant at a 1% level, one and two years after IPO. Three years after IPO, Model (1) and Model (2) do not show an impact. The Chi-squared test shows that these coefficients of Model (1) and Model (2) are not statistically different. This suggests that the SRC rule has a positive effect on
Assets one and two years after IPO, which depends on the size of the firm prior to IPO.
The confidence interval for the combined effect on Assets in Model (2) is presented in Figure 4, panel B. In comparison to IPOs not affected by the SRC rule, one year after IPO Assets increase at least 40 per cent for firms with revenue smaller than $8.9 million prior to IPO. Two years after IPO, the effect is even stronger. There is a total increase of at least 60 per cent in Assets for firms with at most $6.77 million in annual revenue prior to IPO. Three years after IPO there is no impact on Assets. The positive impact on Assets and MarketCap could be explained by a positive impact on
CF/Assets, shown in Table 5.4. The results show a significant negative coefficient at a 5% level one
and two years after IPO for Size x D1 in Model (1) and at a 1% level for Size x D1 x D2 in Model (2). However, the Chi-squared test does not confirm equality between the coefficients in both models. As Model (1) is biased, the results of Model (2) are more reliable. According to Model (2), there is a positive effect of the SRC rule on profitability (CF/Assets) one and two years after IPO, which depends on the size of the firm prior to IPO.
Figure 4 panel C shows the confidence interval of combined effect on CF/Assets for Model (2). In comparison to IPOs not affected by the SRC rule, one year after IPO CF/Assets increases by at least 7 percentage points for firms with at most $11.7 million revenue prior to IPO. Two years after IPO, more firms are significantly affected, but the effect is almost similar. For firms with at most $12.2 million in revenue prior to IPO CF/Assets increases by at least 8 percentage points. Three years after IPO the SRC does not impact CF/Assets.
Regressions for Investment (Table 5.2) and Sales/Assets (Table 5.3) as real outcome variable are also performed to explain the impact of the SRC rule on MarketCap. These performance measures cannot explain the impact of the SRC rule on MarketCap. The results show that going public with reduced costs and loosen legal requirements do not have an impact on Investment and Sales/Assets. Indicating that IPOs under the SRC rule do not create a higher MarketCap and profitability because of increased efficiency, but probably because of increased productivity.
To measure the impact of the SRC rule on job creation, additional regressions for Employees as real outcome variable are performed. The regression results are presented in Table 5.5. The results show that the SRC rule has a positive effect on Employees, one and two years after IPO. The effect
depends on the size of the firm prior to IPO; small SRCs are more affected. This is shown in in the table. One and two years after IPO the coefficient for Size x D1 for Model (1) is negative and statistical significant at 1% level and the coefficient for Size x D1 x D2 in Model (2) is negative and statistically significant at least at a 10% level. The coefficients in Model (1) and Model (2) are not statistically different, measured with the Chi-squared test. Model (2) suggests that there is no effect that impacts everyone equally; D1 x D2 is insignificant one, two and three years after IPO.
Figure 4 panel D presents the combined effect on Employees. The effect of the SRC rule on
Employees has less statistical power; therefore a 90% confidence interval is created. In comparison to
IPOs not affected by the SRC rule, one year after IPO for firms with at most $107 thousand in revenue prior to IPO, the number of Employees at least doubles. Two years after IPO, the effect is even stronger, the number of Employees at least more than doubles for firms with at most $107 thousand in revenue prior to IPO. There is no impact of the SRC rule on Employees three years after IPO. The findings provide evidence that going public early with reduced costs and loosen legal requirements increases Employees one and two years after IPO.
5.1.2 SRC rule: Financial Outcomes
The impact of the SRC rule on the financial side of the firm is also tested. The results indicate that an early IPO with reduced costs and loosen legal requirements decreases leverage and increases stock liquidity. The results suggest that there is no impact on M/B (Table 5.6).
The results show that the SRC rule has a negative impact on Leverage, one and two years after IPO. The regression results, presented in Table 5.7, show a positive and significant coefficient at a 5% level for the variable of interest Size x D1 for Model (1), one and two years after IPO. The coefficient for the variable of interest, Size x D1 x D2 in Model (2) is insignificant. While Model (1) is biased the results are still reliable, because the Chi-squared test confirms equality between the coefficients for the variables of interest of Model (1) and Model (2). Model (2) does not have enough statistical power. Therefore the combined effect cannot be presented, as Model (1) is not able to present a change in the relation independent on the size of the firm prior to IPO. The findings suggest that an early IPO with reduced costs and loosen legal requirements has a negative effect on Leverage one and two years after IPO. The effect depends on the size of the firm prior to IPO.
The results for Model (2) indicate that two years after IPO everyone affected by the SRC rule experience a positive effect on Leverage, unrelated to the size of the firm prior to IPO. The results show that the coefficient for D1 x D2 is significant at a 1% level two years after IPO. Therefore for some firms, the combined effect on Leverage two years after IPO could be positive, depending on the size of the firm prior to IPO.
The results of Liquidity as dependent variable show that the SRC rule has a positive impact on
Liquidity, two and three years after IPO. The regression results presented in Table 5.8 show a negative
three years after IPO. Model (2) does not have enough statistical power. Therefore, the combined effect cannot be presented. The results for Model (1) are still reliable, because Chi-squared test confirms equality between the coefficients for the variables of interest for Model (1) and Model (2). The findings provide evidence that an early IPO with reduced costs and loosen legal requirements has a positive effect on the stock liquidity of the company two and three years after IPO. The effect depends on the size of the firm prior to IPO.
5.2 JOBS Act
In Table 6 till 6.8, the results for the JOBS Act setting for Model (1) and Model (2) are presented for the real and financial outcome variables. The set-up of the tables is similar to the set-up of the tables presented in the previous section.
5.2.1 JOBS Act: Real Outcomes
Overall, the results of the JOBS Act setting are similar to the results of the SRC rule setting. Table 6 presents MarketCap as a performance measure. Similar to the SRC rule setting, an early IPO under JOBS Act with reduced costs and loosen legal requirements has a positive effect on MarketCap one, two and three years after IPO. Thereby, similar to the SRC rule setting, Assets (Table 6.1) and
CF/Assets (Table 6.4) are able to explain the impact of the JOBS Act on MarketCap, while Investment
(Table 6.2) and Sales/Assets (Table 6.3) do not explain the impact. Thereby, as the SRC rule, the JOBS Act also has a positive impact on job creation, measured by the level of Employees (Table 6.5).
However, different from the SRC rule setting Model (2) has limited statistical power under the JOBS Act setting. An explanation could be the limited sample for the non-EGC group, shown in Table 1 and Figure 1. Model (1) has enough statistical power, however this model cannot present the combined effect on performance measures dependent and independent of the size of the firm prior to IPO, as Model (1) does not show the impact independent of the size of the firm prior to IPO. While Model (1) is biased, the results are still reliable, because according to the Chi-Squared test the coefficients for the variables of interest Size x D1 in Model (1) and Size x D1 x D2 in Model (2) are not statistically different.
Another difference is the magnitude. The positive effect of going public in the SRC rule is stronger than in the JOBS Act. The reason of the reduced effect of the JOBS Act could be that small firms in the JOBS Act setting might already have reduced disclosure requirements under the SRC rule.
Different from the SRC rule setting, the JOBS Act also shows a positive effect on Assets three years after IPO (Table 6.1). This indicates that the positive effect of the JOBS Act is present for a longer time period. An explanation could be that the reduced costs and loosen legal requirements of the JOBS Act are more extended than the SRC rule. Thereby, firms can probably maintain EGC status for a longer period, as the revenue cut-off is higher.
For CF/Assets (Table 6.4) as dependent variable, Model (2) shows a negative and significant coefficient at least at a 10% level for D1 x D2, one and two years after IPO. This indicates that EGCs
going public after the JOBS Act are adversely affected, independent on the size of the firm prior to IPO. However, Model (1) shows a positive impact one and two year after IPO, dependent on the size of the firm prior to IPO. Thus, the combined effect depends on the size of the firm prior to IPO. 5.2.2 JOBS Act: Financial Outcomes
Similar to the SRC rule setting, the JOBS Act does not have an impact on M/B (Table 6.6). However, the JOBS Act also does not have an impact on Leverage, shown in Table 6.7. This is different from the SRC rule setting. The reason could be that a different group of firms under the JOBS Act is affected than under the SRC rule. The cut-off for the JOBS Act is higher than for the SRC rule. Therefore, larger firms are affected that might be more mature and might already have a more stable capital structure. This is suggested by a lower standard deviation of Leverage (Table 4). Smaller firms go public under the SRC rule, due to the lower cut-off. These firms might be less mature and might have a less stable capital structure, suggested in Table 3 with a higher standard deviation of Leverage.
Regressions for Liquidity as financial outcome variable are also performed. The results are presented in Table 6.8. Similar to the SRC rule setting, the JOBS Act setting shows a positive effect of the JOBS Act on Liquidity. However, different from the SRC rule setting, under the JOBS Act, the effect is already present one year after IPO, while not three years after IPO. There seems to be a delayed effect for the SRC rule, while under the JOBS Act it is present immediately. Similar to the previous explanation, the difference in size of the affected firms could play a role. It could also be due to macro-economic circumstances, as the SRC rule came into force during the financial crisis, while the JOBS Act came into force a few years after the financial crisis.
5.3 Discussion
The results of the regressions for the SRC rule setting and the JOBS Act setting are very similar and provide evidence that going public early with reduced costs and loosen legal requirement positively influences firm performance and job creation at least until two years after IPO. The positive effect is stronger in the SRC rule setting and among small-targeted firms. This suggests that deregulations are probably successful in achieving their intended goals of economic growth and job creation.
I find a significant positive effect of going public early with reduced costs and loosen legal requirements on MarketCap one, two and three years after IPO. This positive effect on MarketCap could be explained by a significant positive effect on Assets and CF/Assets, while Investment and
Sales/Assets do not explain the impact. The results suggest that these improvements mainly stems
from an increase in productivity instead of efficiency. The regressions results also show a significant positive effect on Employees. For the financial side of firms, both settings show a significant positive effect on Liquidity. Only the results of the SRC rule show a significant negative effect on Leverage.
Significant increases in Assets are identified, which is in line with the research of Mikkelson et al. (2009). Investment is not able to explain the impact on MarketCap. This might indicate that firms have incentives to remain small, which is in line with the research of Gao et al. (2009). Thereby, it
suggests that financing investments are not the main goal of going public, which is contradicting with the research of Rydqvist and Högholm (1995) and Chemmanur et al. (2009).
The results present evidence that going public early with reduced costs and loosen legal requirement has a positive effect on job creation, which supports the evidence of Weild and Kim (2009), Kenney et al. (2012), Ritter (2014) and Borisov et al. (2015).
The evidence for the financial side of the firm is in line with the research of Rydqvist and Högholm (1995) and Pagano et al. (1998), as significant decreases in leverage are identified. Thereby, the identified significant increases in stock liquidity is in line with the research of Brau et al. (2003), while not in line with the research of Cheng et al., (2013). However, Cheng et al. (2013) focuses on public firms instead of IPOs and make a distinction between voluntary and mandatory disclosures.
I find that especially small SRCs and EGCs firms are affected, which is in line with the research of Dambra et al. (2015) and Zimmerman (2015). Thereby, the evidence extends the statement of Chaplinsky et al. (2015), as according to my study a reason of going public under the JOBS Act besides reduced costs could be that it positively affects firm performance and job creation.
Additionally, the results show that small firm IPOs experience a positive effect of deregulation. This is not in line with the claim of Dambra et al. (2015) and the evidence of Mikkelson et al., (1997), Jain and Kini (1999) and Gao et al. (2013) and Davidoff Solomon and Rose (2014) on small IPO performance. However, these studies do not make use of the unique setting of reduced costs and loosen legal requirements.
It should be noted that the results could be positively influenced, as the data collection could suffer from survivorship bias, which could be an issue as the performance improves. Additionally, firms are free to choose any, all, or none of the reduced disclosure requirements. Therefore, the treatment assignment might be endogenous, due to self-selection. However, according to Chaplinsky et al. (2015) the fast majority of IPOs selected EGC status. Thereby, according to Zimmerman (2015), especially small companies make use of the reduced disclosure requirements. This is in line with my research, as small companies are the ones mostly affected.
The findings suggest that the governmental interventions of reducing costs and loosing legal requirements for small issuers are successful in improving performance, and thus economic growth and job creation. Policies that lower costs and legal requirements by easing disclosure requirements have measurable positive implications. However, a limitation of this thesis is that the control group is selective. The control group does not include firms that decided not to go public. It seems that high performing firms stay private due to high costs and legal requirements. I do not examine whether early IPOs with reduced costs and loosen legal requirements perform better compared to private companies. According to Jain and Kini (1994), going public has a negative effect on operating performance. Mikkelson et al. (1997) support this view; they argue that this could be due to the size and age of the firm. Therefore, future research should examine whether the deregulations have a positive effect on performance of public offerings compared to private firms.
