SME Financing Data Initiative
sme-fdi.gc.ca
Common menu bar links
First
Previous
Initial Public Offerings: Status, Flaws and Dysfunctions
1. Overview and Comparisons
1.1 Introduction
Before going into an in-depth look at Canadian primary offerings (initial public offerings, or IPOs), it is necessary to define a few terms and look at certain problems involving definitions and measurements. A number of studies have been conducted in Canada in recent years. They are reviewed, then the main conclusions of US and international studies are examined.
1.1.1 Problems with measurements and definitions
An IPO is a process whereby a company raises capital by issuing shares to investors and subsequently becoming listed on a stock exchange. This is the most conventional way of making the shift from private company to open or public company. There are other ways, such as reverse takeovers or mergers with already listed companies, but these are not discussed here. Primary issues involve complex procedures that are adjusted as necessary along the way, potentially affecting amounts, prices and the issues themselves. Some are withdrawn, but remain on the official lists. Name and status changes are very common, especially for CPCs and small issues. Various sources will therefore often report different data, which explains the lack of exhaustive studies.Footnote 3 In conducting our study, data were systematically checked, but in a number of cases it was impossible to obtain all of the information about share issues or track the shares issued. Some of our analyses thus deal with subsets of the starting sample, which results in bias, especially for the period preceding the introduction of SEDAR.
The presence of CPCs and the very limited size of many conventional issues lead to some difficult problems in terms of international comparisons. Most US databases list only issues of shares priced at or over $5, likely as a result of US regulations. The 1996 National Securities Market Improvement Act (NSMIA) distinguishes between "covered securities" and securities that are "not covered securities." The former come under the SEC. These are conventional issues that lead to the listing of the securities on national markets like the NYSE or NASDAQ. Securities that are "not covered securities" come under the states and are governed by Rules 504 and 505 of Regulation D as well as Rule 147 for intrastate offerings. Offerings of this type can, however, involve amounts of up to US$20 million. American studies are based solely on offerings under the SEC. It is, however, possible to identify a large percentage of US issues of shares priced between $1 and $5. We did this for some years.
It is virtually impossible to identify US issues of shares valued at under $1, which come under state regulations and specific programs like SCOR (Small Capital Offering Regulation). In Canada, more than three out of four offerings fall within this category. The Canadian primary issues market is thus largely comparable to the local US issues market and comparisons have to take that into account.
Moreover, even though they are officially listed as IPOs, some issues are hard to view as primary share issues in the conventional sense, as they stem from demutualizations or privatizations, or are conducted by big business subsidiaries created for the purpose. We also look at foreign issues of shares in Canada and Canadian issues abroad. We end with a brief description of the Capital Pool Company program, which accounts for nearly half of all primary issues in recent years.
What is an IPO?
The official list of IPOs in Canada is published by the Financial Post.Footnote 4 However, this list covers offerings by mutual funds, trust companies and subsidiaries as well as independent operating companies. Offerings arising from demutualizations or floated by subsidiaries of existing companies are also listed. These offerings use various investment vehicles. It is thus advisable to define the IPOs warranting in-depth analysis.Footnote 5
For the purposes of this study, IPOs are operations leading directly to an increase in the equity capital of an operating company or CPC. Mutual fund issues are therefore not included, since the funds they raise are used to acquire securities already issued. For the same reason, we leave out trust company offerings. This exclusion has major effects on the total amounts listed, as trust companies alone raised nearly $20 billion between 1991 and 2000. Land trusts (REITs) are also excluded, along with some rare venture capital company offerings.
Included in the study are offerings by subsidiaries, which often involve significant amounts. Since a subsidiary is majority owned by an already listed company, such offerings are not initial offerings for the group. This applies, for example, to Bell Canada International, 74% owned by BCE and responsible for one of the biggest share issues of the period ($466 million in 1997). These transactions have been included, since they are not dealt with separately in studies conducted in other countries. However, these offerings may behave very differently from other issues.
While almost all studies show that primary offerings have an abnormally negative performance in the medium term, Cusatis et al. (1993) highlight a strongly positive pattern in the case of offerings by subsidiaries. The accumulated yield over market after three years is 33.6%, whereas this figure tends to be very negative for offerings as a whole. The relatively major weight of offerings by subsidiaries in Canada may skew the results of empirical analyses by lowering the negative performance of securities in IPOs.
Individual securities
A number of investment vehicles are used for IPOs. The most widely used is the common share, but various other types are in general use and dealt with here as follows.
- Issues of restricted voting shares provide limited or no voting privileges. These have been included with issues of common shares, as they do not differ in terms of residual equity. Preferred share issues offer holders the right to receive a fixed dividend before any dividends are paid on common shares, and preferred shareholders are also entitled to a portion of the company's residual assets if it is wound up. IPOs of preferred shares are highly infrequent and accounted for in this study as a separate item in the "other" category.
- Flow-through shares issued by oil, gas and mineral exploration companies allow for certain tax deductions for qualifying exploration and development costs that can "flow through" from the company to shareholders. These offerings are grouped in the "other" category with units and warrants. The warrants included in units give the bearer the right to acquire a fixed number of further securities at a set price for a fixed period of time. Offerings of limited partnership units represent interests in a partnership consisting of a general partner, who manages the partnership, and limited partners, who provide the investment capital. The liability of the limited partners is limited to the value of their initial investment, provided they do not become involved in management. Limited partnerships generally invest in a specific industry sector (like real estate or oil and gas) and often bring some tax benefits that can "flow through" from the partnership to the limited partners. They have not been included with IPOs, and are also left out of the US studies.
- Trust unit offerings represent interests in the net assets and net income of a trust. These trusts are set up to invest in real property (real estate investment trusts), royalties from oil or gas production (royalty trusts) or the income generated from one or more businesses (income trusts). Since the proceeds of these issues do not directly increase the companies' capital, we do not consider them IPOs. The same goes for investments made by mutual funds. The treatment given these various classes of securities is recapped in Appendix 1.
Demutualizations
Demutualization is a process by which eligible mutual insurance company policyholders become shareholders in a company with common shares.Footnote 6 In 1999 and 2000, the following five mutual life insurance companies were demutualized and appear in our lists:
| Name | Issue Date | Gross Proceeds ($) |
|---|---|---|
| Canada Life Financial Corp. | 04-11-99 | 523,250,000 |
| Clarica Life Insurance Company | 21-07-99 | 951,377,366 |
| Industrial-Alliance Life Insurance Company | 10-02-00 | 339,806,250 |
| Manulife Financial Corporation | 30-09-99 | 1,566,381,088 |
| Sun Life Financial Services of Canada | 27-03-00 | 1,166,773,675 |
We have considered these companies separately in our analysis since, although seemingly initial public offerings, they actually involve an exchange of interests in a mutual company for shares, without any net capital contribution. The gross proceeds of demutualization issues are very high compared with average proceeds from conventional IPOs and have to be dealt with properly to avoid distortions.
Privatizations
Offerings arising from privatizations also have a major influence on the data, in that they are large in size, yet do not provide any new capital for the privatized corporations. Privatization involves a transfer of activities from the federal, provincial or municipal public sector to the private sector. Canadian governments began to privatize their corporations in the mid-1980s, and in 1995 the federal government floated an initial public offering to privatize the shares of the national railway. According to Levac and Wooldridge (1997, p. 29),Footnote 7 to obtain a higher selling price that included a premium for acquiring control of the enterprise, most privatizations in Canada have taken the form of direct buyouts by existing private businesses rather than IPOs. We were able to identify only five offerings arising from privatization between 1991 and 2000, for total gross proceeds of $4,690.01 million. These were:
| Name | Issue Date | Gross Proceeds ($) |
|---|---|---|
| Petro-Canada | 1991-06-25 | 546,062,192 |
| Cameco Corporation | 1991-07-11 | 120,000,000 |
| Nova Scotia Power Inc. | 1992-07-29 | 851,346,660 |
| Canadian National Railway | 1995-11-28 | 2,262,600,000 |
| Manitoba Telecom Svc | 1997-01-07 | 910,000,000 |
We have kept these privatizations separate for the purposes of our analysis, as they did not yield any new capital for the corporations.
For purposes of this study, the amounts included in gross proceeds from share offerings represent the total raised through the offerings, regardless of the country of residence of the buyer. Offerings conducted totally in the US by Canadian businesses are merely recorded, whereas offerings conducted simultaneously in both countries are included in the sample. Securities sold in Canada through issues by US companies are also included.
Offerings under the Capital Pool Company program
A large number of Canadian primary issues come under the CPC program. These issues differ substantially from the traditional IPO. The first CPC program, then known as the Junior Capital Pool, was launched in Alberta in November 1986 by the Alberta Securities Commission and Stock Exchange. In 1997, the British Columbia Securities Commission and Vancouver Stock Exchange launched a similar program, known as the Venture Capital Pool. The current CNDX program was started on March 1, 2000 to replace the two earlier ones following the merger of the Vancouver and Alberta stock exchanges in November 1999. It is the product of a joint effort by the Alberta, Saskatchewan, ManitobaFootnote 8 and British Columbia securities commissions and the CDNX.Footnote 9 The program was then taken up by Ontario and Quebec, undergoing changes each time. According to CDNX Policy 2-4,Footnote 10 the objective of the CPC program is to provide businesses with an instrument for speedier financing than the traditional IPO route.
This program allows for the creation of a shell company (capital pool) with a stock exchange listing without any operating track record or assets other than cash, for the sole purpose of identifying and acquiring small private companies. Once this "qualifying transaction" has been completed, the new entity can be listed on the exchange under the regular rules. The CPC offerings therefore correspond to the shell company's listing on the stock exchange. There are various restrictions on these offerings, including the following for the period that interests us:
- The amounts raised must be not less than $200,000 and not more than $500,000; the minimum price per share is $0.15;Footnote 11 directors and senior officers must contribute $100,000 to $500,000 in start-up capital; this start-up capital plus the amount raised through the IPO cannot exceed $700,000; this money must be used exclusively to identify and complete a transaction that will result in a listing on an exchange;
- The only business permitted to be undertaken by the shell company is the identification and evaluation of assets or businesses with a view to completing a transaction. It cannot own assets other than cash;
- The directors must possess expertise in managing publicly held businesses, this being considered the prime asset of the capital pool.
Offerings under the CPC program are included in this study, but analyzed separately due to their very special characteristics. Their medium-term performance is not examined, owing to the major distortions resulting from the "qualifying transactions."Footnote 12
1.1.2 Canadian research on primary issues
Canadian studies on primary issues are relatively scarce and tend to focus on the problem of initial underpricing, as shown by Jog and Riding (1987), Suret et al. (1990), Jog and Srivastava (1994) and Jog (1997). According to Jog and Riding (1987), short-term performance following the initial listing of a security was 11.5% in the period 1971-1983. Suret et al. (1990) place average underpricing at 12% for 86 Ontario primary issues in the period 1979-1985, though underpricing was not seen in 63 issues eligible for the Quebec Stock Saving Plan (RÉAQ) during this period, probably as a result of associated tax benefits. Jog and Srivastava calculate average underpricing of 5.67% for the period 1984-1992. Jog extends the study by Jog and Srivastava (1994) to cover offerings in 1993-1994, noting average underpricing of 7.89%. He stresses that underpricing of primary issues in Canada seems to be declining. Between 1971 and 1983, 62% of primary issues were underpriced, compared with 47% for the period 1984-1992. According to Jog and Srivastava and Jog, there is no need for concern by Canadian decision-makers about the phenomenon of underpricing of primary issues or its potential impact on companies' plans to make public offerings. However, these studies look only at offerings listed on the Toronto Stock Exchange.
Falk and Thornton (1992) place adjusted average initial underpricing at 19% for primary issues on the Toronto Stock Exchange between 1983 and 1988, 25% for those listed on the Montreal Exchange and 307% for issues listed on the Alberta Stock Exchange, including a number of CPCs. The major differences between offerings in the various jurisdictions justify the broader analysis being carried out here. The report prepared by Groupe SECOR Inc. for the Task Force on the Future of the Canadian Financial Services Sector (SECOR, 1998) does not point to any problem with the operation of the primary issues market in Canada.
However, the report does stress a structural lacuna relating to equity capital for amounts under $1 million.Footnote 13 MacIntosh (1994) shows the significant costs of IPOs, especially for small issues, and produces some evidence that Canadian companies use primary issues less readily than US companies. Admittedly, the studies cited deal mainly with the 1970s and 1980s. Here again, a study of recent activity is needed.
1.1.3 Summary of major studies conducted worldwide
A review of the countless studies conducted in the US on various aspects of primary issues is beyond the compass of this study and so we will limit ourselves to highlighting the main findings. Syntheses do exist, for example by Ritter (1998) and Ritter and Welch (2002) for the US market and by Jenkinson and Ljungqvist (2001) for other markets.Footnote 14 The main findings of the research on IPOs can be organized into four major topics. The first is the issuing activity itself, especially its cyclical nature. The second is the cost of issues, especially the initial underpricing, which is a major factor. The second part of this study looks at this phenomenon and reviews earlier research. The third topic is medium-term performance and, more recently, the survival of new issues. Studies on this topic are discussed in the third part of this paper. Lastly, an emerging component deals with issuing mechanisms.
Issuing activity
Primary issue activity in the various markets is highly cyclical, with "hot" periods succeeding "cold." In hot periods, offerings abound and prices are relatively high, but there is severe initial underpricing as well, keeping investor demand strong. Medium-term underperformance, it seems, is also more serious in these periods. The most plausible explanation for this situation raises questions about the thinking of investors who, in times of stock exchange euphoria, take a special interest in primary issues.
Business managers apparently use financing strategies that are strongly influenced by market fluctuations. They tend to issue shares when they have seen upward movement in the market. In general, however, IPOs have become much more frequent since the early 1980s. Fama and French (2002) note that the pace of new business listings in the US rose from 140 in the late 1970s to nearly 600 in the late 1980s.
Studies of issuing activity also highlight the growing share of issues by technology companies, which represented 25% of IPOs in the early 1990s, compared with 37% after 1995 and 72% during the Internet bubble. According to Ritter and Welch, the percentage was 29% in 2001.
The significant increase in the proportion of IPOs by money-losing companies was another striking feature of the 1990s, rising from 19% in the 1980s to 79% in 1999-2000. It still stood at 49% in 2001.
Fama and French (2002) ascribe the major drop in survival rates to the lower quality of new listings. Because of poor performance, the likelihood of survival after 10 years has risen from 14.4% to 40.2% of primary issues [sic – TR]. These values, however, are derived solely from issues followed by listings on major markets (NYSE, NASDAQ, AMEX) and the writers acknowledge that this limitation skews their results. Although the sample is limited to relatively large issues by Canadian standards, there is clearly a strong connection between size, profitability and survival rate. Issues by small businesses have less chance of surviving and companies in this category are much less profitable on average than the big companies.
Finally, multinational studies of issuing activity help us to more effectively situate primary issue activity in Canada. Ljungqvist and Wilhelm (2002) counted 2,861 IPOs in 15 European countries, or 260 a year, for 15 markets, including the main European ones. On average, a primary issue can raise $131 million in Germany, $74 million in France and $93 million in the United Kingdom. The period 1995-1999 saw an average of 585 US IPOs at issue prices above $1 and 404 large issues (at $5 and over) per year. The average amount raised through each of these issues was $84 million. The US IPO market alone is twice the size of the market for all 15 countries of the European Union.
Issuing mechanisms
According to a number of researchers, including Loughran et al. (1994) and Chowdhry and Sherman (1996), the mechanisms used to value and distribute securities in primary issues affects the stock exchange listing process. The most widespread approach in the US and Canada is book building, in which the lead broker gathers information on the buying intentions of potential institutional and individual investors during "road shows" and establishes a demand curve based on a number of criteria. Biais and Faugeron (2000) and Sherman (2001) emphasize that book building is preferable to the auction approach. They are of the view that book building can be seen as a dynamic auction conducted by brokers, with the advantage that the brokers can freely allocate securities to reward investors who reveal their buying plans. Cornelli and Goldreich (2001) also stress that preference is given to institutional investorsFootnote 15 that are ready to buy and hold securities in their portfolios for the long term.
The issuing price of shares is set so that there is apparent surplus demand and distribution is left to the underwriter's discretion. The book building mechanism is often used for large firm-commitment issues. Kutsuna and Smith (2000) stress that, since book building was introduced in Japan in 1997, Japanese issuing firms have preferred this approach to the auction approach introduced in 1989.
Ljungqvist et al. (2001) also confirm that the international introduction in the 1990s of book building as a system for valuing and allocating securities has done much to improve efficiency in primary issue markets. However, Ritter (1998) points out that book building has not escaped criticism. Loughran et al. (1994) emphasize that brokers prefer book building, but express doubt as to whether this mechanism is in the best interests of the issuing company.
1.2 Canadian stock issues
1.2.1 Overview by class of issues
Table 1 shows issuing activity for all sectors in Canada, including the companies that listed through the CPC program. The data cover a 10-year period from 1991 to 2000, for which we identified a total of 1,891 Canadian issues, including 10 issues of shares arising from the demutualization of insurance companies or privatization of public corporations.
There were 1,023 traditional offeringsFootnote 16 for total gross proceeds of nearly $32.03 billion. These traditional offerings include issues of common shares and other issues, including issues of units, preferred shares and flow-through shares. However, common share issues were the most common (775), raising more capital ($29 billion) than the remaining 248 issues ($3 billion).
The CPCs created between 1991 and 2000 accounted for 868 issues for total gross proceeds of $223 million. The number of offerings under the CPC program exceeded the number of initial public offerings of common shares. Because of CPC program terms and conditions, both the total amount issued and the average ($257 K) are clearly much lower than the amounts involved in traditional primary issues ($31 million). The Canadian market is characterized by very large numbers of new small businesses. On average, primary issues raise $131 million in Germany, $74 million in France and $93 million in the UK.
In Canada, the average amount initially raised through an IPO is $17 million taking all issues together and $31 million if CPC issues are excluded. Canadian primary issues are thus more numerous, but their gross proceeds trail those in other countries: on average, there are 189 IPOs a year in Canada, compared with 47 in France, 80 in the UK and 43 in Germany. In the US, the average amount raised through conventional offerings ($5 and over) is around C$82 million during the period in question.
Canada is clearly a small issue market. If demutualizations and privatizations are excluded, average gross proceeds rise to $2.5 million. This value is further increased by a few offerings by subsidiaries.
Table 1 also shows the annual distribution of issues under the CPC program, of common shares outside this program and of other classes of securities, including share units, preferred shares and flowthrough shares. Issues of fixed-income securities are excluded, along with issues by mutual funds, trusts and limited partnerships. Gross proceeds (GP) are expressed in millions of dollars.
1.2.2 Main features
Geographical features
Table 2 provides a breakdown of primary issues by province of incorporation. Over 90% of the issues were by companies incorporated in one of the four provinces of Ontario, British Columbia, Alberta and Quebec. Most primary issue activity was centred in those four provinces, which in July 2001 accounted for over 85% of Canada's population.
| Province | 1991 | 1992 | 1993 | 1994 | 1995 | 1996 | 1997 | 1998 | 1999 | 2000 | Total |
|---|---|---|---|---|---|---|---|---|---|---|---|
| AB | 7 | 11 | 51 | 64 | 55 | 65 | 86 | 65 | 30 | 58 | 492 |
| BC | 3 | 5 | 17 | 16 | 15 | 31 | 22 | 47 | 55 | 211 | |
| F | 1 | 4 | 2 | 4 | 1 | 12 | |||||
| ON | 3 | 3 | 11 | 10 | 14 | 16 | 27 | 11 | 5 | 100 | |
| OTH | 1 | 1 | 2 | 4 | 1 | 5 | 3 | 4 | 6 | 27 | |
| QC | 2 | 2 | 6 | 1 | 6 | 6 | 3 | 26 | |||
| Total | 7 | 18 | 61 | 100 | 89 | 101 | 143 | 123 | 98 | 128 | 868 |
| Province | 1991 | 1992 | 1993 | 1994 | 1995 | 1996 | 1997 | 1998 | 1999 | 2000 | Total |
|---|---|---|---|---|---|---|---|---|---|---|---|
|
Sources: Financial Post, Report of New Issues, Cancorp Financials, www.sedar.com and www.tsx.com/. Province of incorporation of issuing company: QC: Quebec, ON: Ontario, AB: Alberta, BC: British Columbia, OTH: Nova Scotia, Manitoba, Saskatchewan, New Brunswick, Newfoundland, Northwest Territories, Prince Edward Island, Yukon, F: Issues by companies headquartered outside Canada. |
|||||||||||
| AB | 11 | 15 | 103 | 92 | 72 | 95 | 138 | 88 | 40 | 70 | 724 |
| BC | 28 | 25 | 39 | 44 | 42 | 54 | 77 | 46 | 69 | 77 | 501 |
| F | 7 | 2 | 3 | 12 | 11 | 13 | 13 | 5 | 4 | 1 | 71 |
| ON | 11 | 12 | 42 | 57 | 36 | 52 | 66 | 55 | 40 | 28 | 399 |
| OTH | 4 | 4 | 2 | 5 | 7 | 2 | 12 | 7 | 6 | 11 | 60 |
| QC | 0 | 4 | 19 | 9 | 11 | 24 | 21 | 15 | 12 | 21 | 136 |
| Total | 61 | 62 | 208 | 219 | 179 | 240 | 327 | 216 | 171 | 208 | 1891 |
Sixty issues (3%) originated in eight other provinces [and territories? – TR]. Finally, there were 71 issues by foreign companies, mainly American, accounting for 4% of the total.Footnote 17
Most issues associated with a CPC program were from Alberta (57%), the province where the program originated, and British Columbia (24%).
Table 2 shows the distribution of initial public offerings in Canada, 1991-2000, by province of issuing company's head office.
Issue size
Table 3 shows the breakdown of initial public offerings in 1991-2000 by size. Issues with gross proceeds of $5 million or less account for 61% of all issues, but total only $812 million, or less than 3% of the total. More than 67% of the total capital raised was from issues with gross proceeds of over $100 million, demutualizations and privatizations. Most of the funds raised through primary issues in Canada over the 10-year period thus came from 59 very large issues, or 6% of total offerings. This makes comparison with the situation in the US difficult. Fewer than 20 Canadian issues a year (184 in all) raised more than the US$20 million threshold and are thus comparable to issues of SEC-covered securities under the NSMIA (1996). All others (90%) could have met regional regulations and been included in "not covered securities" within the meaning of the NSMIA. The Canadian IPO market is therefore basically comparable to the US market for local offerings.
Sectoral characteristics
Table 4 shows that 41% of the traditional issues in Canada are by resource companies (mining and oil and gas) basically concentrated in two provinces. Over half (51%) of the issues in British Columbia are by mining companies. This province accounted for 58% of all issues by mining companies in Canada in 1991-2000. Alberta accounts for 79% of oil and gas issues and these offerings represent more than half of all offerings by Alberta companies.
Ontario accounts for over 55% of issues by financial services companies, more than half of consumer goods issues and 39% of technology issues. In Quebec, 35% of the issues are by technology firms and 26% come from the manufacturing and pharmaceutical sectors.
In short, most issuing companies in British Columbia and Alberta are small resource companies. Conversely, many Ontario issues are by very large financial services concerns. Company characteristics thus vary widely from province to province.
1.2.3 "Hot" issuing periods
The number of primary common share issues varies greatly from year to year, as shown in Figure 1. Activity was low in 1991 and 1992 from the standpoint of numbers of issues and gross proceeds and can thus be described as "cold," compared with the "hot" issue market of 1993-1994. This pattern was repeated in the years that followed.
Figure 1 shows Canadian primary issues occurring in waves that crested in 1993-1994 and 1996-1997, 1997 being the record year for this activity. The concentration of primary issues in certain periods is consistent with the window of opportunity hypothesis advanced by Ritter (1991): companies choose to become listed when markets are "hot," in order to maximize the proceeds of offerings.
Description of Figure 1
Figure 2 shows the differences in US primary issue activity. For example, 1996 saw heavy issue activity, in contrast with 1991 and 1998,Footnote 18 and 1999 was a "hot" year in the US market, due in large part to the arrival on the stock market of various dot.coms, while the Canadian issues market remained relatively quiet.
Description of Figure 2
Some phenomena analyzed below, such as initial underpricing and long-term under-performance, seem to be tied directly to the cyclical nature of the primary issues market.
1.2.4 Creation and disappearance of Canadian businesses
From 1991 to 2000, the number of companies listed on Canadian stock exchanges declined by 5% overall, from 4,342 to 4,124,Footnote 19 despite 1,891 initial public offerings. In all, in this period, 1,218 issues raised capital of $1 million or less, mainly in association with the CPC program (868 issues). The net loss of 195 companies over this period reveals a high "mortality rate" possibly resulting from mergers, reprivatizations or delistings.Footnote 20 Each year saw the disappearance of almost as many public corporations as were created.
This observation is consistent with the findings of Fama and French (2002), which highlight the frequent disappearance of the securities of newly listed US "small" businesses. Since Canadian issues are distinctly smaller than similar offerings in the US, it is not surprising that their survival rate is well under the US rate. The scope of this study does not allow us to arrive at any conclusions concerning the 10-year survival rate, but the fact that more companies are going under than coming up suggests an extremely low survival rate.
1.2.5 Technology offerings
These companies are in the CIS 25 technology sub-group when their main sectoral code is on the list in Appendix 2. The only problem industry for classification purposes is biotechnology, where the same code covers drug manufacturers and companies geared mainly to R&D. To be consistent with practice, we have not included these companies under sub-sector 25, but under subsector 28, "pharmaceuticals."
Table 5 shows primary issues activity by technology firms varying year over year. Activity was heaviest in terms of numbers of issues in 1995, 1996 and 2000. The proportion of initial public offerings by technology companies was highest in 2000, with over 44% of total issues and nearly 50% of gross proceeds from issues. Ontario is very active in the area of technology offerings (39%), followed by British Columbia (22%) and Quebec (16%).
| Province | 1991 | 1992 | 1993 | 1994 | 1995 | 1996 | 1997 | 1998 | 1999 | 2000 | Total |
|---|---|---|---|---|---|---|---|---|---|---|---|
|
Sources: Financial Post, Report of New Issues, Cancorp Financials, www.sedar.com and www.tmx.com. Province of incorporation of issuing companies: QC: Quebec, ON: Ontario, AB: Alberta, BC: British Columbia, OTH: Nova Scotia, Manitoba, Saskatchewan, New Brunswick, Newfoundland, Northwest Territories, Prince Edward Island, Yukon, F: Foreign companies. When companies have federal charters, the province of incorporation is determined by head office location. |
|||||||||||
| AB | 1 | 2 | 1 | 3 | 5 | 12 | 3 | 2 | 2 | 31 | |
| BC | 3 | 2 | 3 | 2 | 9 | 14 | 6 | 2 | 4 | 10 | 55 |
| F | 1 | 3 | 2 | 4 | 3 | 3 | 2 | 18 | |||
| ON | 4 | 15 | 9 | 9 | 20 | 12 | 10 | 7 | 10 | 96 | |
| OTH | 1 | 3 | 2 | 6 | |||||||
| QC | 1 | 6 | 1 | 1 | 7 | 5 | 5 | 2 | 10 | 38 | |
| Technology sector total | 3 | 9 | 26 | 17 | 24 | 50 | 41 | 23 | 17 | 34 | 244 |
| Technology sector GP ($ millions) |
2.1 | 172.6 | 620.9 | 272.3 | 373.1 | 759.7 | 2187.1 | 538.2 | 422 | 1852.5 | 7200.5 |
| Total for all sectors | 54 | 44 | 147 | 119 | 90 | 139 | 184 | 93 | 73 | 80 | 1023 |
| GP for all sectors ($ millions) |
1271.1 | 1561.9 | 3752.20 | 3541.5 | 2954.2 | 2617.9 | 6001.4 | 2072.7 | 4433.3 | 3827.9 | 32034.4 |
| % of technology companies | 5.56 | 20.45 | 17.69 | 14.29 | 26.67 | 35.97 | 22.28 | 24.73 | 23.29 | 42.50 | 23.85 |
| % GP of technology companies | 0.17 | 11.05 | 16.55 | 7.69 | 12.63 | 29.02 | 36.44 | 25.97 | 9.52 | 48.39 | 22.48 |
1.3 Share offerings in Canada and the US
1.3.1 Problems with measuring US activity
Tables 6 and 7 allow a comparison of common share issues in Canada and the US, with separate figures for the technology sector. The number of issues listed is misleading: official US lists exclude securities issued at under US$5, whereas all issues are considered in Canada. Shares worth less than $5 are viewed as highly speculative as they do not meet the minimum criteria for NASDAQ listing and are not regulated in the same way.
| Year | Total | Technology Stock Offerings3 | |||
|---|---|---|---|---|---|
| #1 | #2 | GP1 | # | GP | |
|
"#" = number of issues; GP = gross proceeds in US$ billions.
1 Data from Ritter (2003), available at: http://bear.cba.ufl.edu/ritter/work_papers/IPOs2002.pdf. Data exclude issues priced at US$5 and under, ADRs, best efforts, units and Regulation A offers, REITs, partnerships and closed end funds.
|
|||||
| 1991 | 288 | - | 15.77 | - | - |
| 1992 | 397 | - | 22.20 | - | - |
| 1993 | 507 | - | 29.26 | - | - |
| 1994 | 416 | - | 18.3 | - | - |
| 1995 | 465 | 834 | 28.87 | 308 | 10.66 |
| 1996 | 666 | 1142 | 42.48 | 460 | 24.27 |
| 1997 | 484 | 555 | 33.22 | 267 | 15.53 |
| 1998 | 319 | 422 | 35.11 | 168 | 14.67 |
| 1999 | 490 | 557 | 65.46 | 362 | 35.9 |
| 2000 | 385 | - | 65.68 | - | - |
| Total | 4417 | 3510 | 356.35 | 1565 | 101.03 |
We have tried to adjust the US data for the years 1995-1999 by including issues of securities priced between $1 and $5. It is virtually impossible to get information about issues priced under $1 in the US. Most of these issues are governed by separate regulations and no conventional prospectuses are produced.
To perform a proper comparison of issues in the two countries, it is necessary to exclude Canadian issues priced at under $1. This leaves 419 Canadian issues at prices higher than $1. This number does not change much when the threshold is set at $2.Footnote 21 The gross proceeds of issues vary little when issues at $1 and under are excluded: the total for all issues is reduced by only $376 million.
1.3.2 General comparisons
The amounts raised in Canada total C$28.47 billion, whereas issues in the US brought in US$356.35 billion, the equivalent of C$404 billion if the amounts raised every year are adjusted for the average exchange rate for that year. Canadian issues therefore represent 7% of those in the US, which is far lower than the proportion of US GDP represented by Canada's GDP.Footnote 22
The number of US issues is greatly understated, as issues priced under US$5 are not reported in Column 1 of Table 7. To obtain an idea of the number of US issues, we attempted to complete the US lists for 1995-1999. The ratio between the total number identified and the number of issues priced at over $1 is 1.45:1 (3510 / 2424), and so the total number of initial public offerings in the US may be estimated at around 6,400. Based on these figures, the number of Canadian issues represents 6.6% of US issues – once again, far less than the Canada-US GDP ratio.
1.3.3 Technology stock offerings
Technology stock offerings total $6.97 billion in Canada, or 24.45% of the overall capital raised. Technology issues in the US and their gross proceeds were identified for 1995-1999, making it possible to arrive at the comparison in Table 7.
In the US, technology issues raised US$101 billion, or C$144.3 billion. The $6.97 billion raised in Canada is therefore about 4.83% of the US amount. Canada clearly lags far behind the US in terms of financing technology companies through share issues.
This situation is no different if issues by Canadian companies made completely in the US are included, since the number is too small to influence the results.Footnote 23
Public financing of technology companies is therefore very modest in Canada, compared with the situation in the US. This is paradoxical in view of the heavy activity reported by Canadian venture capital companies. In 1999, for example, the Canadian Venture Capital Association reported more than 1,200 investments in technology sectors and managed funds were in excess of $12 billion. Canadian venture capital companies raised $8.4 billion in 1995-1999, or 160% of the amount raised by primary issues – and this holds true for every year studied except 1997.
1.4 Conclusion
Our analyses show that primary issues are far fewer in Canada than in the US and the capital raised is appreciably less after standardization by GDP. The differences between the two countries are even greater if issues not officially listed in the US, that is, issues by small businesses equivalent to most Canadian offerings, are included. Canada seems to be a market of very small stock offerings. However, the inclusion of privatization and demutualization offerings may partially obscure this phenomenon, as these issues account for a major proportion of the total gross proceeds raised in the 1990s. Issues by subsidiaries of already listed companies also account for significant amounts, further reducing the measurable issuing activity of completely new firms. Paradoxically, big corporations do not seem to make frequent use of the primary issue market.
The gap between primary issues in Canada and those in the US is even more pronounced in the field of technology. And it is even wider than first appears in that the main Canadian technology stock offerings are by subsidiaries (Bell, AT&T) and not by new companies.
In the following sections, we look at three possible explanations. It may be that issues are rare because they are too costly: an inefficient market (Part 2) might be lowering the supply of issues. It is also possible that primary issues are scorned due to the poor performance of new securities. This would curb demand by institutional and private investors (Part 3). Lastly, it may be that acts and regulations, policies and institutional behaviours are responsible for the gap. These factors are the subject of the final part of this study.
Footnote 3 This phenomenon is not limited to Canada. American writers have a great deal of difficulty producing consistent figures. Figures for US issues differ appreciably in Ritter (1998), Ritter (2003) and Ritter and Welch (2002). For 1996, for example, the 1998 article reports 845 IPOs, while the 2003 article reports 666 (Table 5) and the 2002, 621 (Table 1).
Footnote 4 Financial Post, Record of New Issues, Toronto, 1991-2000.
Footnote 5 It is also important to have our definitions match the ones generally used in US studies, to make comparison possible. The offerings excluded here are also excluded by Ritter (1998).
Footnote 6 http://www.sunlife.com
Footnote 7 The fiscal impact of privatization in Canada
Footnote 8 The Winnipeg Stock Exchange joined the CDNX in November 2000. It had previously been running a similar scheme called the Keystone Company program.
Footnote 9 A CPC can be initiated by residents of another province, but the CPC shares cannot initially be acquired by anyone who is not a resident of one of these four provinces.
Footnote 10 http://www.tmx.com/en/pdf/Apr13-05-CPC.pd.
Footnote 11 The CDNX Capital Pool raised the maximum issue size from the $300,000 set by the two earlier programs to $500,000, and the minimum issue price from $0.10 to $0.15.
Footnote 12 Inasmuch as the CPC have no commercial activity of their own, it would be more appropriate for the issue conducted on completion of the "qualifying transaction" to be an IPO.
Footnote 13 This study seems to be limited to big business and significantly underestimates the number of primary technology issues in Canada.
Footnote 14 Initial Public Offerings Resource page. Emphasis on studies, research resources, and experts.
Footnote 15 Jog (1997) states that issuers agree that institutional investors play a major role in initial share movements but continue holding the shares for longer than private investors. As shown by Boehmer et al. (2002), however, the institutions get a proportionately larger share of the "good" issues, with higher long-term yields. There is probably a connection between this longer holding period and the advantage they seem to enjoy in initial share allocation.
Footnote 16 By "traditional offerings," we mean ones not associated with the CPC program.
Footnote 17 48 of the 59 traditional IPOs were by companies headquartered in the United States and the 11 others were by companies in Bermuda (1), Indonesia (1), Israel (1), Mexico (1), Hong Kong (2), Ghana (1) and Europe (4).
Footnote 18 These primary issues seem to have been mainly in specific sectors, such as fashion in 1996 and 1997. The primary issue by the Gucci fashion house was behind a number of other primary issues by fashion houses like Donna Karen (June 1996) and Ralph Lauren (June 1997).
Footnote 19 Sources: Montreal Stock Exchange: Monthly Review, 2000; Statistics, Research and Market Information (1991), Toronto Stock Exchange Review, Alberta Stock Exchange Review, Vancouver Stock Exchange Review and CDNX Monthly Review. ftp://ftp.cdnx.com/Publications/CDNXReviews/.
Footnote 20 In December 1998 alone, 25 securities were delisted from the Toronto Stock Exchange (TSE Review, December 1998) and 4 were delisted from the Montreal Stock Exchange.
Footnote 21 Securities pricing habits in Canada are different from those in the US, where values of $100 to $200 are relatively common. They are rare in Canada. We therefore considered the US$5 threshold as equivalent to $1 or $2 in Canada.
Footnote 22 Based on 1998 data, Canada's GDP is 10.9% of the US figure (Statistique Québec).
Footnote 23 Issues by Canadian companies made outside Canada are basically made in the US and are relatively rare. Jog and Hitsman (2000) found 21 between 1994 and 2000, and 11 companies conducted simultaneous issues in the US and Canada. They note that using the US market confers no tangible benefit on issuing companies. Initial underpricing and post-issue price volatility are more marked in the US than in Canada.
Information on Downloading a PDF Reader
To access the Portable Document Format (PDF) version you must have a PDF reader installed. If you do not already have such a reader, there are numerous PDF readers available for free download or for purchase on the Internet:
