Unit IPOs: a mechanism for small businesses to go public.

The ability to raise capital to finance growth projects is a major challenge to new and growing small businesses which are typically private companies. Their efforts to raise capital on favorable terms are constrained by the lack of liquidity of private shares and by the absence of generally

available information on their future prospects. Banks and other lenders seek higher premiums for the lack of liquidity for non-traded shares and often impose highly restrictive covenants on private companies to account for the information asymmetry and the resulting moral hazard and agency problems. The problem of information asymmetry is particularly acute for small businesses since there is generally less information available on them. Problems with private financing have caused entrepreneurs to turn to the initial public offering (IPO) market to raise capital. A firm enhances its liquidity by going public and simultaneously is able to convey more information on its future prospects by virtue of increased market monitoring. However, by going public, a firm incurs direct costs in the form of underwriter and administrative fees and indirect costs in the form of underpricing.

Private firms planning to become public corporations have to identify both the timing and the appropriate mechanism to make this transition. Usually, firms go public through a sale of common stock, however, another alternative is to sell units, which are a combination of common stocks and common stock purchase warrants sold as a package. Numerous academic studies have examined stock IPOs but relatively little is known about unit IPOs.(1)

This study attempts to identify whether unit IPOs represent a feasible mechanism for small businesses to make the transition from private to public ownership. Specifically an attempt is made to address the following questions: (1) Are unit issuers primarily small busineses? (2) Are the overall transaction costs higher for unit IPOs than stock IPOs?(2) (3) Are there any benefits to issuing unit IPOs instead of stock IPOs?

The results of this study suggest that unit IPOs are issued primarily by smaller, younger, riskier firms with little or no prior operating history or sales. The median total transaction cost of going public as a percentage of the gross proceeds for unit IPOs is 3.40 percent greater than the median transaction cost for similar stock IPOs. Advantages of unit issues include higher proceeds at the IPO and the ability to trigger a subsequent round of financing without incurring the additional cost of a seasoned offering. These advantages are particularly attractive to smaller, riskier, growing businesses characterized by high levels of uncertainty.

PREVIOUS RESEARCH

Prior studies in the IPO literature have focussed primarily on the short-run underpricing behavior and the presence of "hot and cold" markets. One of the earliest prominent studies to document significant IPO underpricing was by Ibbotson (1975). The existence of the "hot issues" phenomena, which represents periods in the economy characterized by severe levels of IPO underpricing, was documented by Ibbotson and Jaffe (1975) and Ritter (1984). Subsequently studies by Ibbotson, Sindelar, and Ritter (1988), Carter and Manaster (1990), and Ritter (1991) confirmed the continued existence of severe underpricing in IPOs.

Only two studies have attempted to measure the transaction costs of going public and, in both instances, the analysis was confined to stock IPOs. Ritter (1987) found that firm commitment IPO issuers surrendered a staggering 22 percent of the market value of their equity during the process of going public. Aggarwal and Rivoli (1991) provided similar findings.

Several theoretical studies have attempted to provide an equilibrium explanation for the persistence of IPO underpricing. Rock (1986) suggested that underpricing results from the "winner's curse" problem due to asymmetrically informed investors. Allen and Faulhaber (1989), Grinblatt and Hwang (1989) and Welch (1989) suggested that issuers deliberately underprice to signal their quality. In their models, underpricing is used as a mechanism to resolve the information asymmetry problem between issuers and investors. Although these models provide intuitively appealing explanations for the underpricing phenomena, strong empirical support has been lacking.(3)

Recent efforts have been aimed at documenting the long-run investment performance of IPOs. Ritter (1991) examined a sample of IPOs from 1977-1984 and found that these firms substantially underperformed similar firms over a three-year window from the date of going public. Loughran and Ritter (1993) confirmed the earlier finding of significant IPO underperformance and also attempted to measure the determinants of IPO volume. They found that while the level of stock prices has significant explanatory power in explaining IPO volume, the growth rate in gross national product (GNP) is insignificant. Thus, their results suggested that changes in IPO volume are not caused by changes in growth opportunities. They suggested that a "timing effect" exists so that firms have a disproportionate chance of going public near market peaks.

Despite the abundance of IPO literature, the problems of small businesses have received relatively little attention. Young and Zaima (1986) examined the relationship between underpricing and variables such as age of firm, type of underwriter contract and industry effect in the context of small business IPOs. Johnson and Miller (1985) examined how IPO offering methods are matched to businesses. Several questions, however, persist in the context of small business IPOs, and the current study attempts to address some of these issues.

ADVANTAGES OF UNIT IPOs

To understand the advantages of unit IPOs, it is useful to examine their basic institutional details by looking at an example. The initial public offering made by Network Imaging Corporation in April 1992 represents a typical unit offering. The company offered 1.5 million units, each consisting of one common stock and one redeemable common stock purchase warrant. The units were priced at $4.00 each and the warrants entitled the holder to exchange each warrant for a common stock at a price of $5.00 during the first 30 months and $6.00 during the next 30 months from the date of the initial offering. In general, warrants provide investors with an opportunity to make significant gains if the stock price escalates while limiting their losses to the cost of the warrant if the stock price declines.

Why do some firms go public with units instead of a straight stock offering? The inclusion of warrants complicates the process of going public. Unless offsetting benefits are obtained, issuers prefer the typical stock IPO. There are several advantages of unit IPOs which may appeal to smaller, growing businesses that are perceived as being risky by investors. A unit offering provides higher gross proceeds at the IPO for a given fraction of the firm sold in comparison to a stock IPO. The additional proceeds come from the sale of the warrants attached to the unit. However, the higher proceeds at the IPO can be offset by losses if the stock price of the issuer escalates beyond the exercise price of the warrants and investors exercise their options.

Another advantage of unit IPOs is that equity financing is raised in two stages with the second stage being conditional on firm performance as measured by an increase in its stock price. The sequential nature of equity financing permits unit IPOs to reduce some of the typical problems of information asymmetry in the IPO market. Due to lack of operating history or market prices, investors are unable to accurately value new issues. As a result, low quality issuers have an incentive to paint themselves as attractive candidates and sell their stock at inflated prices. Subsequently, when market prices are established, investors realize that they received overpriced shares but meanwhile the issuers have realized their gains. Over a period of time investors drop out of the IPO market due to persistent negative returns and a market failure occurs, preventing even fairly priced IPOs from raising capital. Recently, Allen and Faulhaber (1989), Grinblatt and Hwang (1989), and Welch (1989) have developed theoretical models which posit that high value issuers overcome this type of informational problem by signalling their quality with underpricing at the IPO and raising equity in two stages. The cost of underpricing is offset by gains achieved through higher prices at the seasoned offering. The ability to underprice and wait for a seasoned offering to complete their financing needs provides an opportunity for high value issuers to distinguish themselves from low value issuers.

Unit IPOs are another form of two-stage financing and could be used as a signalling mechanism by high value issuers. Initially the issuer completes partial financing by selling a given number of units at the IPO. Subsequently, if the firm succeeds and the stock price exceeds the exercise price of the warrants, further capital is raised by exercise of the warrants. By issuing units, issuers are signalling their willingness to raise further equity only if the firm performs well. The ability to wait until the firm succeeds to raise financing provides high value issuers with a mechanism to separate themselves from a pool of issuers and to help reduce the information asymmetry problem.

Finally, the units provide a means for smaller, riskier firms with a greater degree of uncertainty surrounding their prospects to successfully complete the IPO. For example, if the initial demand indications during the "road show" or pre-marketing stage are unexpectedly low for a stock IPO, the issuers can throw in the warrants and convert the issue into a unit IPO to increase investor interest. The risk-reward content of warrants may appeal to a certain class of investors who hope to make significant gains if the stock price escalates. Further, their downside risk is bounded since, in the worst case situation, their loss is confined to the cost of the warrant. Thus, investors who otherwise may not have been interested in the stock IPO are drawn to participate in the unit IPO due to the potential of making significant gains.

The advantages of unit IPOs seem particularly suited for smaller, start-up businesses which are perceived by investors as riskier propositions in comparison to larger more established ventures. For instance, larger size IPOs attract considerable attention from investment bankers, institutional investors, and analysts, which leads to the release of information on their future prospects. These firms by virtue of reduced information asymmetry are able to raise sufficient capital to satisfy their needs at the IPO itself. The smaller issuers, plagued by problems of information asymmetry, need to raise funds in stages. The staged financing feature of unit offerings, without the cost of a seasoned offering, is particularly attractive to smaller businesses since it allows them to offset the problem of information asymmetry and provides a low cost means to approach the equity markets more frequently.

If units are attractive to smaller, growing businesses why do some firms issue stocks instead of a unit IPOs? The answer lies in the transaction cost of unit versus stock issues. If issuing units leads to higher transaction costs in comparison to stock IPOs, the issuers have to consider the trade off between the benefits of issuing units versus additional transaction costs. These issues are examined in the remainder of this article.

DATA DESCRIPTION AND METHODOLOGY

The sample was drawn from the population of all firm commitment IPOs issued during the period 1977-1984. The sample excludes penny stock issues priced below $1 and regulation A securities.(4) Foreign issuers and American depository receipts (ADRs) were also excluded. These restrictions resulted in a final sample of 1,457 IPOs. The details of these 1,457 IPOs were obtained from Going Public: The IPO Reporter and Investment Dealers Digest's Five Year Directory of Corporate Financing. Additional data were obtained by examining individual prospectuses and the S1, S2 or S18 reports filed by issuers with the SEC prior to the IPO.

The overall sample was divided into sub-samples consisting of unit and stock IPOs. Units represented approximately 14 percent of the sample (203 issues). The differences between unit and stock IPOs on variables such as age of firm, prior sales history, book value of equity, risk of issue, size of offering, etc., are compared using the Mann-Whitney nonparametric statistic. The sales history is measured by the most recent 12-month revenues prior to going public. The book value is defined as the tangible book value of equity prior to the IPO. The risk of the issue is measured by the standard deviation of returns spanning days 2-31 after the IPO.(5)

For each of the 1,254 stock IPOs, the initial returns are calculated in the usual manner by using the first closing bid price and the offer price at the IPO. For the 203 unit IPOs, the initial unit price and the closing unit price at the end of the first day of aftermarket trading are used to compute the initial returns which are as follows: I[R.sub.i] = ([U.sub.1] - [U.sub.0])/[U.sub.0], where IR represents the initial returns and [U.sub.1] and [U.sub.0] are the closing bid unit price on the first day of trading and the unit offer price, respectively. In order to avoid the problem of skewed distributions, the median initial returns are used to estimate the location parameter. Skewed distributions are driven by outliers which can seriously bias the inferences. Further, the bias is magnified when comparing the difference in central tendency between two skewed groups of unequal sample sizes. For skewed distributions, Winkler and Hays (1975) strongly recommended the use of medians as a measure of central tendency and hence, median values are used throughout this paper.

The total cost of going public has two components: (1) initial returns at the IPO, and (2) direct costs of underwriting and other expenses. The direct costs are measured as the sum of the underwriter commissions and other expenses which include legal, printing, and auditing fees. The transaction costs are measured as a percentage of the gross proceeds raised at the IPO. Finally, to estimate the determinants of the transaction costs of going public, ordinary least square (OLS) regressions are estimated. The independent variables include the log of offer size of the firms, risk, age, and a dummy variable to represent the type of IPO (unit versus stock).

EMPIRICAL RESULTS

In table 1, the characteristics of unit and stock IPOs are compared. The unit IPOs are younger firms with a median age of three years in comparison to a median age of seven years for stock IPOs with the difference being statistically significant at the 1 percent level. The median IPO offer size for unit firms is $3.6 million in comparison to a median offer size of $8.25 million for stock IPOs with the difference being statistically significant at the 1 percent level. Further, unit IPOs are characterized by significantly lower levels (at 1 percent) of book value of equity and sales in comparison to stock IPOs. The book value of equity is another proxy for the firm's ability to generate funds internally. A firm with larger book equity prior to the IPO is less likely to have a subsequent issue.

Table 1 COMPARISON OF UNIT AND STOCK IPOs

                                                      Mann-Whitney
Description                    Stock IPOs  Unit IPOs  Statistic
Age of Firm                    7 years     3 years    -9.05(a)
Gross Proceeds Raised at IPO   $8.25M      $3.60M     -11.37(a)
Initial Price Per Share or
Unit                           $10.0       $5.8       -11.92(a)
Annual Sales Prior to Going
Public                         $13.01M     $0.41M     -13.80(a)
Pre-Offer Book Value of
Equity                         $2.91M      $0.30M     -12.71(a)
Initial Returns                1.47%       1.78%      0.14
Risk of the Issue              0.026       0.040      6.59(a)
Number of Firms                1254        203        -

Note: Median values of variables reported for unit and stock IPOs. The Mann-Whitney Statistic measures whether the difference between unit and stock IPOs on each of the variables is significant.

(a)Significant at the 1 percent level.

These results provide strong evidence to suggest that unit IPOs are primarily start-up companies with little or no operating history or sales prior to the IPO and are reliant on external financing making them more likely to have a subsequent issue. The unit IPOs are also characterized by higher levels of uncertainty as seen by the standard deviation of aftermarket returns. The median standard deviation of aftermarket returns is equal to 0.04 for unit IPOs in comparison to 0.026 for stock IPOs, and the difference is statistically significant at the 1 percent level. However, there is no statistically significant difference in initial returns between the two groups with a median initial return of 1.78 percent for unit IPOs and 1.47 percent for stock IPOs.

In table 2, the IPO issuers are classified by gross proceeds and type of issue. The largest proportion of unit IPOs is in the lowest gross proceeds class ($1.50-$3.99 million) where they represent 31 percent of all IPOs. It appears that unit IPOs are preferred by smaller firms since approximately one-third of issuers raising less than $4 million at the IPO choose to go public with a unit issue. As the amount raised at the IPO increases, the percentage of unit IPOs decreases.

[TABULAR DATA OMITTED]

In table 3, the median initial returns segmented by gross proceeds for both unit and stock IPOs are reported. As shown earlier in table 1, there is no statistically significant difference in initial returns between the two types of IPOs. However, the comparison in table 1 was made without controlling for differences in offer size. The analysis in table 3 provides an opportunity to compare the initial returns of unit IPOs and stock IPOs of similar size. The results suggest that except for offers that raise between $4.00-$9.99 million, the median initial returns for unit IPOs are either not different from or are statistically significantly lower than that of stock IPOs.

Table 3 INITIAL RETURN DISTRIBUTION OF UNIT AND STOCK IPOs

                     Unit IPOs              Stock IPOs
                            Median                 Median
Gross Proceeds              Initial                Initial
(millions of                Returns                Return
dollars)         Number     (percent)   Number     (percent)
 1.50-3.99        112          0.00       259         0.00
 4.00-9.99         71          4.16       443         1.56(a)
10.00-29.99        12          0.00       392         1.25
Above 30.00         8          1.51       160         2.36(a)
All Offers        203          1.78      1254         1.47

Note: A Mann-Whitney test measures whether the difference in median values between unit and stock IPOs is significant for each of the size categories.

(a)Significant at the 1 percent level.

(b)Significant at the 5 percent level.

In table 4, the median direct expenses as a percentage of gross proceeds are reported for both unit and stock IPOs. There is clear evidence of economies of scale for both types of IPOs. Both the median underwriter commissions and other administrative expenses decrease with the amount raised at the IPO for both types of issues. However, there are greater economies of scale in other administrative expenses in comparison to underwriter commissions. Further, the results show that for an offer of any given size the direct costs of going public are higher for unit IPOs in comparison to stock IPOs. Apparently the inclusion of the warrants in unit issues leads to higher underwriting and other administrative costs.

Table 4 MEDIAN DIRECT EXPENSES AS A PERCENTAGE OF GROSS PROCEEDS FOR UNIT

AND STOCK IPOs

                             Panel A
               Median Direct Expenses for Unit IPOs
                                         Other           Total
Gross Proceeds   Number of  Underwriter  Administrative  Direct
(millions of     Unit       Discount     Expenses        Costs
dollars)         IPOs       (percent)    (percent)       (percent)
1.50-3.99        112        10.00        9.04            19.04
4.00-9.99         71        10.00        7.08            17.08
10.00-29.99       12         7.80        3.70            11.50
Above 30.00        8         7.00        1.40             8.40
                             Panel B
               Median Direct Expenses for Stock IPOs
                                         Other           Total
Gross Proceeds   Number of  Underwriter  Administrative  Direct
(millions of     Stock      Discount     Expenses        Costs
dollars)         IPOs       (percent)    (percent)       (percent)
1.50-3.99        259        10.00        7.33            17.33
4.00-9.99        443         8.00        5.00            13.00
10.00-29.99      392         7.11        2.47             9.58
Above 30.00      160         6.75        1.13   7.88

In table 5, the median total transaction costs as a percentage of the gross proceeds are reported for both types of issues. The total transaction cost is calculated as the sum of the direct costs and the cost of underpricing. The evidence suggests a negative relationship between IPO offer size and the total transaction cost of going public for both the unit and stock IPOs. However, the negative relationship is not monotonic for stock IPOs since the total transaction cost increases for issues above $30 million. It is also clear from table 5 that the total transaction cost of going public with units is higher than that of stock IPOs except for issues raising above $30 million. Since there are only eight unit firms in this category, no strong conclusions can be drawn.

Table 5 MEDIAN TOTAL TRANSACTION COSTS AS A PERCENTAGE OF GROSS PROCEEDS

FOR UNIT AND STOCK IPOs

                               Panel A
             Median Total Transaction Costs for Unit IPOs
                             Median      Median      Median
Gross Proceeds   Number of   Direct      Initial     Total
(millions of     Unit        Costs       Returns     Costs
dollars)         IPOs        (percent)   (percent)   (percent)
1.50-3.99        112         19.04       0.00        19.04
4.00-9.99         71         17.08       4.16        21.24
10.00-29.99       12         11.50       0.00        11.50
Above 30.00        8          8.40       1.51         9.91
                               Panel B
             Median Total Transaction Costs for Stock IPOs
                             Median      Median      Median
Gross Proceeds   Number of   Direct      Initial     Total
(millions of     Stock       Costs       Returns     Costs
dollars)         IPOs        (percent)   (percent)   (percent)
1.50-3.99        259         17.33       0.00        17.33
4.00-9.99        443         13.00       1.56        14.56
10.00-29.99      392          9.58       1.25        10.83
Above 30.00      160          7.88       2.36        11.24

In table 6, the results of OLS regressions estimating the determinants of the total transaction cost of going public are reported. The independent variables include a dummy variable to represent the type of offering (unit versus stock), the log of the offer size, the age of the firm and a measure of firm risk.(6) In regression 1, the only independent variable is the type of offering which is represented by a dummy variable taking on the value 1 if the issue is a stock IPO and 0 otherwise. The coefficient of the type variable is -0.1865 and is statistically significant at the 1 percent level which suggests that the total transaction costs of unit IPOs is higher than stock offerings without controlling for differences in size and risk of the issues. In regression 2, the regressions are estimated after controlling for differences in offer size. Both the type of IPO and the size of the issue are statistically significant variables in estimating transaction costs of an IPO. The results in regression 3 suggest that younger firms incur higher transaction costs, and the negative relationship between the age of the firm and transaction costs is statistically significant at the 1 percent level. Finally in regression 4 the difference in transaction costs between unit and stock IPOs is compared after controlling for differences in size, risk, and age of the firms. The coefficient of the variable type is - 0.034 and is statistically significant at the 5 percent level. Hence, the total transaction costs of unit IPOs is 3.40 percent higher than that of similar stock IPOs.

Table 6 DETERMINANTS OF TRANSACTION COSTS (TC) OF GOING PUBLIC SUMMARY OF

OLS REGRESSION RESULTS T[C.sub.j] = [[beta].sub.1] Type + [[beta].sub.2] Log Size + [[beta].sub.3] Risk + [[beta].sub.4] Age

                  Regression Number
Factor Description  1           2           3           4
Intercept           0.4356      1.03        0.242       0.628
Type = 1 if Stock
IPO;                -0.1865     -0.11       -0.07       -0.034
= 0 if Unit IPO     (-6.08)(a)  (-3.80)(a)  (-5.03)(a)  (-1.69)(b)
Log Size                        -0.074                  -0.054
                                (-8.95)(a)              (-9.55)(a)
Risk                                                    1.80
                                                        (6.99)(a)
Age                                         -0.0013     -0.0008
                                            (-3.95)(a)  (-0.24)
[R.sup.2]           .0246       .0752       .030        .131
F Value             36.99       59.63       24.44       52.10

Note: t-statistics are reported within parenthesis.

(a)Significant at the 1 percent level.

(b)Significant at the 10 percent level.

This study also examines whether the transaction costs of unit and stock IPOs exhibit an industry effect.(7) In table 7, the median total transaction costs as a percentage of gross proceeds for both unit and stock IPOs are segmented by industry. Firms are classified into industries based on three digit Standard Industrial Classification (SIC) codes. The IPOs were classified into 11 major industry groups and the results indicate weak industry effect. For stock IPOs, the largest number of issues occurred in the computer and data processing industry (249), followed by communication equipment and services (195) and financial services industry (77). The total transaction costs as a percentage of proceeds for stock IPOs ranged from a low of 10.82 percent for the drug and pharmaceutical industry to a high of 15.92 percent for the medical instruments industry. For unit IPOs, the largest number of issues occurred in the communication equipment and services industry (38), computer and data processing (27), and business services industry (22). The total transaction costs as a percentage of gross proceeds varied from a low of 16.35 for the chemicals industry to a high of 27.16 percent for the financial services industry.

Table 7 BREAKDOWN OF TOTAL TRANSACTION COST OF GOING PUBLIC BY INDUSTRY

AND TYPE OF IPO

                                 Unit IPOs       Stock IPOs
                               Total                   Total
                               Transaction             Transaction
                    Number of  Costs(*)     Number of  Costs(*)
Industry            Firms      (percent)    Firms      (percent)
Oil and Natural
 Gas                 8         21.58         75        12.73
Drugs and
 Pharmaceuticals    11         20.31         54        10.82
Computer and Data
  Processing        27         21.03        249        11.14
Communication
 Equipment and
 Services           38         18.10        195        13.80
Medical
 Instruments        11         19.09         60        15.92
Bus Services        22         19.92         42        14.15
Air Transportation   3         19.87         24        12.23
Financial Services  11         27.16         77        11.25
Chemicals            3         16.35         12        11.73
Machine Equipment
 and Supplies        8         25.20         38        12.03
Others              61         19.37        428        12.08

(*)As percentage of gross procees raised at the IPO.

In summary, the smaller, younger, riskier businesses attempt to raise capital to finance their growth with unit IPOs while stock IPOs are favored by larger, well-established companies with a track record of prior sales. The total transaction costs of going public are higher for the unit IPOs in comparison to similar stock IPOs. Clearly small businesses need to consider the tradeoff between higher transaction costs against the benefits of unit IPOs in making their choice of going public with a stock versus unit IPO.

DISCUSSION AND CONCLUSION

This study provides evidence to suggest that unit IPOs are a popular mechanism employed by younger, smaller businesses that have higher levels of information asymmetry surrounding their future prospects. Practically one-third of issuers raising less than $4 million at the IPO go public with a unit issue. Larger, established companies that have a significant prior operating history and sales record prefer to go public with a stock IPO. Further, the total transaction costs of going public are higher for unit IPOs in comparison to stock IPOs. The addition of warrants to the stock issue increases the underwriting cost and other administrative expenses.

The results documented in this study provide several managerial implications especially for small, growing businesses. The IPO market is clearly an important viable alternative source of capital for small businesses and development-stage firms. At lest two major issues facing small business managers during the prepublic planning process include selecting both the timing of the public offering as well as the mechanism in the form of a unit versus stock issue. The timing issue requires managers to determine the appropriate stage of development of the firm at which an initial public offering should be considered instead of venture capital or other sources of private financing. The results of this study suggest that both the age of the firm and size of the offering have a significant impact on the total transaction costs of going public. Thus, managers of small businesses with little or no operating history or sales may find it worthwhile to wait to become more established before going public. This may require potential IPO issuers to initiate several rounds of venture capital financing prior to going public. As a result, when the firm goes public it would be more established and capable of raising higher proceeds at the IPO thereby reducing the transaction costs. Another managerial issue is related to the choice between unit and stock IPOs. Issuers who are perceived by investors as risky, or growing businesses who anticipate multiple offerings to raise capital to finance their growth, might find unit IPOs a more appropriate vehicle to launch a public offering. However, they need to weigh the advantages of unit issues against the disadvantage of incurring higher transaction costs in comparison to stock IPOs. On the other hand, established issuers, characterized by lower uncertainty and the capacity to raise a sizeable amount at the IPO, should prefer a stock issue due to the lower transaction costs.

The benefits of issuing units needs to be subjected to a closer scrutiny. A promising area of future research will be to track the aftermarket performance of unit IPOs to see if additional capital is being raised through the exercise of the warrants. Lower overall transaction costs for a two-stage financing process might provide a satisfactory explanation as to why smaller, growing businesses opt to go public with a unit IPO.

(1)For example, Ibbotson (1975), Ritter (1984), Carter and Manaster (1990) and Ritter (1991). These studies have primarily examined the short-run underpricing and long-run investment performance of stock IPOs.

(2)Ritter (1987) identified two quantifiable cost components of going public: the direct cost of underwriting, printing, and legal fees and the indirect cost of underpricing.

(3)Jegadeesh, Welch, and Weinstein (1991) and Garfinkel (1993) found no evidence to support the signalling models.

(4)Regulation A securities are IPOs raising less than $1.5 million in gross proceeds. These issues are exempt from the usual disclosure requirements by the U.S. Securities and Exchange Commission (SEC).

(5)Ritter (1984) experimented with several ex-ante and expost measures as proxies for risk and found the results to be essentially similar. Since the standard deviation of aftermarket returns has become a widely accepted measure of risk in the IPO literature, it is also employed in this article to mirror firm risk.

(6)Correlation analysis does not indicate any serious potential multicollinearity problem among the independent variables.

(7)Past studies have shown an industry effect in IPOs. For example, Ritter (1984) found that the hot issues market existed only for the oil and gas industry during the 1980-81 period.

REFERENCES

Aggarwal, R., and P. Rivoli (1991), "Evaluating the Costs of Raising Capital Through an Initial Public Offering," Journal of Business Venturing 6(5), 351-360.

Allen, F., and G.R. Faulhaber (1989), "Signalling by Underpricing in the IPO Market," Journal of Financial Economics 23, 303-323.

Carter, R., and S. Manaster (1990), "Initial Public Offerings and Underwriter Reputation," Journal of Finance 45, 1045-1067.

Garfinkel, J.A. (1993), "IPO Underpricing, Insider Selling, and Subsequent Equity Offerings: Is Underpricing a Signal of Quality?" Financial Management (Spring), 74-83.

Going Public: The IPO Reporter and Inventment Dealer's Digest Five Year Directory of Corporate Financing 1980-1984. Philadelphia: Howard and Co.

Grinblatt, M., and C.Y. Hwang (1989), "Signalling and the Pricing of New Issues," Journal of Finance 44, 393-420.

Ibbotson, R.G. (1975), "Price Performance of Common Stock New Issues," Journal of Financial Economics 3, 235-272.

Ibbotson, R.G., and J.F. Jaffe (1975), "Hot Issue Markets," Journal of Finance 30, 1027-1042.

Ibbotson, R.G., J. Sindelar, and J.R. Ritter (1988), "Initial Public Offerings," Journal of Applied Corporate Finance 1, 37-45.

Jegadeesh, N., I. Welch, and M. Weinstein (1993), "An Empirical Examination of Seasoned Equity Offerings and IPO Underpricing," Journal of Financial Economics 34, 153-175.

Johnson, J.M., and R.E. Miller (1985), "Going Public: Information for Small Businesses," Journal of Small Business Management (October), 38-44.

Loughran, T., and J.R. Ritter (1993), "The Timing and Subsequent Performance of IPOs: The U.S. and International Evidence," University of Illinois working paper.

Ritter, J.R. (1984), "The 'Hot Issue' Market of 1980," Journal of Business 57, 215-240.

_____ (1987), "The Costs of Going Public," Journal of Financial Economics 19, 269-281.

_____ (1991), "The Long-Run Performance of Initial Public Offerings," Journal of Finance 46, 3-27.

Rock, K. (1986), "Why New Issues Are Underpriced," Journal of Financial Economics 15, 187-212.

Welch, I. (1989), "Seasoned Offerings, Imitation Costs, and the Underpricing of Initial Public Offerings," Journal of Finance 44, 421-449.

Winkler, R., and W. Hays (1975), Statistics: Probability, Inference and Decision. New York: Holt, Rinehart and Winston.

Young, J.E., and J.K. Zaima (1986), "Does It Pay To Invest in Small Business IPOs?" Journal of Small Business Management (April), 39-50.