1 International evidence on the determinants of the decision to list.

Abstract:

This paper offers empirical insights into the determinants of listing activity using, annual data sampled across 38 exchanges. Listing activity in developed countries stock markets is influenced by a range of variables which capture stock market and business conditions. The single most important factor in explaining the current level of listing activity, however, is the past level of listing activity. For emerging stock markets, none of the variables tested possessed any real explanatory power. Evidence is also found to suggest that high listing activity is accompanied by an increase in seasoned equity issues.

Keywords:

IPO; LISTING ACTIVITY; INTERNATIONAL STOCK MARKETS.

1. Introduction

Corporate finance theory suggests a number of possible reasons that may explain the decision of a company to list on a stock exchange. These include: the need to raise capital to pursue growth opportunities; the desire to enhance the value of the firm; and the creation of a market for ownership of the company. Further, signalling theory suggests that firms may use the decision to list to convey favourable information to the market. Ritter and Welch (2002) provide an overview of this literature. More recently, the results of La Porta, Lopes-de-Silanes, Schleifer and Vishny (1997) and Vitols (2001) suggest that the decision to list may be influenced by the country's legal and banking system.

Each of these different models of the decision to list assumes independence of the firm in the decision making process. The empirical literature has found extensive evidence of cycles in listing activity, however, which directly contradicts this assumption. This clustering phenomenon was first identified in the context of Initial Public Offerings (IPOs) by Ibbotson and Jaffe (1975) and periods of high IPO activity have come to be known as 'hot issue markets' (Ritter 1984). (1) Evidence of this behaviour for the US market is presented in Ibbotson, Sindelar and Ritter (1988, 1994), while international evidence may be found in Loughran, Ritter and Rudqvist (1994).

In response, a strand of literature has emerged which focuses on attempting to identify the determinants of listing activity. For example, Cook, Jarrell and Kieschnick (2003) and Helwege and Liang (2004) test US data and find that investor sentiment has significant explanatory power. Brailsford, Heaney, Powell, and Shi (2000) find that stock market and business conditions explain the cyclical nature of US listings, while Ljungqvist (1997) reaches a similar conclusion for Germany. Lowry and Schwert (2002) find that US IPO cycles are driven by the information revealed during the listing process of past IPOs. Lowry (2003) tests whether the adverse selection costs of issuing equity, the aggregate capital demands of private investors, or investor optimism can explain the variability of US IPO volume and finds support for the latter two determinants. Pagano, Panetta and Zingales (1998) consider the listing behaviour of Italian firms and find that the level of activity is positively related to the stock-market valuation of firms in the same industry.

The purpose of this paper is to provide additional insights into the cyclical behaviour of listing activity. Understanding the motivations which underlie the decision to list is important for financial market researchers, government regulators and stock exchanges. While a study of the time series behaviour of listing activity is not new, this study is unique insomuch as it will test a broad cross section of international listing data. Further, a wide range of variables which the previous literature has found to be important in explaining the nature of listing activity are considered. One such factor is the prevailing stock market conditions. Ritter and Welch (2002) propose a semi-rational symmetric information model for clustering, in which entrepreneurs are slow to adjust their valuation of their own firm and are more inclined to sell after market share valuations have increased (i.e. during a bull market run). Helwege and Liang (2004) and Lucas and McDonald (1990) also argue that firms will time their listing to coincide with a rising market. On a similar theme, a number of studies have suggested that the decision to list is a function of current underpricing (Lowry & Schwert 2002; Ibbotson, Sindelar & Ritter 1988, 1994; Rock 1986; Loughran & Ritter 2002; Brailsford, Heaney, Powell & Shi 2000) or listing activity (see Choe, Masulis & Nanda 1993). (2) This relationship is based on the assumption that underpricing is an indication of improved valuations by the market and firms will take advantage of these 'windows of opportunity' to list (Ibbotson, Sindelar & Ritter 1994; Loughran, Ritter & Rudqvist 1994). Thus, the conditions in the stock market are thought to have an important role to play in explaining the cyclical nature of listing activity.

In addition to stock market conditions, business conditions are a second possible determinant of listing activity (see Brailsford et al. 2000; Ljungqvist 1997). For example, Schulz (2003) models a firm which has the option to invest in a project whose value is positively correlated with the business cycle. The firm does not have sufficient capital to fund this project and so the decision to proceed will necessitate the firm going public. Where a number of firms possess such projects, the value of which are positively correlated to the factors represented by a general business indicator, then the decision to list will also be correlated. Thus, the firm's desire to raise capital through a public offering is thought to be positively related to general business conditions in the home economy.

A third possible determinant of listing activity focuses on bond market conditions. Benninga, Helmantel and Sarig (2003) argued that firms go public when investors place relatively high values on the cash flows of the firms that list. If interest bearing securities have low yields, for example, investors will value the expected cash flows of firms more highly. Thus, individuals may be prepared to pay a premium to secure the rights to company dividends, which provides firms with an incentive to list. On a similar theme, Lerner, Shane and Tsai (2003) argue that where firms cannot access the equity market on favourable terms, they may enter into less favourable financing arrangements, which may include issuing debt. The arguments of Lerner, Shane and Tsai (2003) and Benninga, Helmantel and Sarig (2003) suggest that corporate bond market activity may capture information as to the public demand for new listings.

Using information related to the stock and bond market as well as business conditions, this paper will empirically test the determinants of listing activity for a wide range of international exchanges. The remainder of this study will proceed as follows. Section 2 introduces and describes the data to be used in this study. The list of variables which proxy the determinants of listing activity are also discussed. Section 3 empirically tests the determinants of listing activity and discusses the results as well as some robustness analysis. Finally, section 4 concludes the paper and suggests some directions for further research.

2. International Listing Activity

The World Federation of Exchanges (WFoE) is a private organization which collects market statistics directly from member exchanges using a standardised list of definitions and calculation methods to ensure comparability of data. The WFoE database provides annual estimates of the number of new listings for 38 stock exchanges sampled over the period 1995 to 2002, providing a total of 304 observations. This data captures information on the number of companies which list shares for the first time on a stock exchange and so excludes information on backdoor and cross-listings. This database is a composite of information primarily relating to IPO activity, but also includes a minority of companies that were already public but unlisted.

Grouped by geographic region, these exchanges are: North America (Amex, Mexico, Nasdaq, NYSE, TSX Toronto Stock Exchange), South America (Buenos Aires, Lima, Santiago, Sao Paulo), Europe, Africa and the Middle East (Athens, Copenhagen, Deutsche Borse, Euronext, Helsinki, Irish, Istanbul, Italian Exchange, JSE South Africa, London, Luxembourg, Oslo, Spanish Exchanges, Stockholm, Swiss Exchange, Tel-Aviv, Vienna, Warsaw), and the Asia-Pacific (Australian, Hong Kong, Jakarta, Japan, Korea, Kuala Lumpur, New Zealand, Philippines, Singapore, Taiwan, Thailand). (3) Annual information on the number of domestic companies newly listed on the main and parallel markets for each of these stock exchanges is presented in table 1.

Distinguishing between four general geographic regions, it is clear that global listing activity is dominated by the North American exchanges, excluding Mexico. The average number of listings for the five exchanges in this region peaked in 1996 at 216 new listings per exchange. The NASDAQ generated the highest number of listings each year in North America even after the bursting of the dot.com bubble where the average fell from 510 (1995-2000) to 122 (2001-2002). Across all other regions, only the London Stock Exchange exhibits a level of listing activity over the entire sample period which is comparable to the US exchanges with a minimum of 161 listings in 1999 and a maximum of 366 listings in 2000. The Spanish Exchanges (Barcelona, Bilboa, Madrid and Valencia) experienced a substantial increase in listing activity after 1999 with over 500 listings each year (the average prior to this was only 87). Beyond the London and Spanish exchanges, the remaining stock exchanges rarely exceeded 100 listings per year and a number of exchanges failed to reach double figures during the sample period (Buenos Aires, Irish, Luxembourg and Vienna). In South America, Lima and Santiago experienced a decline in listing activity throughout the sample period however, Sao Paulo exchange experienced a doubling of listing activity in 1998. The European experience is quite diverse with some exchanges (Copenhagen Ireland, Luxembourg, Austria) rarely exceeding ten listings per annum. Other European exchanges (Deutsche Bourse, Euronext, London and Spain) however, exhibited a level of listing activity in some years which would rival that of the NYSE. (4)

A examination of the time series properties of these data reveals clear evidence of heightened periods of listing activity. Figure 1 presents a plot of the average number of new listings by geographic region. At this aggregate level and over a relatively modest 8 year sample period, listing activity is clearly quite variable. Further evidence may be found by considering the individual exchange data presented in table 1. There was a flurry of listing activity on the Japanese exchanges from 2000 on (182 listings per year on average), which was over double the average level of listing activity over the previous five years (79 listings on average). This variability of the number of listings is also a feature of the data for the smaller exchanges. For example, 1996 and 1997 generated relatively high levels of listing activity on the Mexico exchange (14 and 18 issues respectively) relative to the rest of the sample period where only five or fewer listings took place. While the dataset is not long enough to infer the existence of cycles in listing activity, these two examples highlight the variable nature of listing activity which is a feature of the entire dataset. Further, the correlation between individual stock market listing activity is typically low and often negative which suggests that these heightened periods of listing activity are typically not synchronised, especially across geographic regions. For example, the correlation between the total number of listings in each of the four regions indicates that listing activity on the South American exchanges is closely correlated to listing activity in North America (0.754) but negatively to the Europe, Africa, Middle East (-0.099) and Asia Pacific (-0.233) regions. On the other hand, a positive correlation is observed between listing activity in the Asia Pacific region and Europe, Africa and the Middle East (0.409).

[FIGURE 1 OMITTED]

3. Theoretical Determinants Of Listing Activity

The discussion in section 2 indicates that the level of listing activity within an individual country and across particular geographic regions exhibit a number of stylised characteristics. It is an interesting empirical question to consider the extent to which these patterns can be explained and the remainder of this article shall address this question.

The existence of time-varying 'windows of opportunity' to list are thought to be created by favourable stock market conditions (see Lucas & McDonald 1990; Ritter & Welch 2002; Helwege & Liang 2004). One obvious metric which captures the existence of these windows is the past level of IPO underpricing. (5) Unfortunately, a time series of underpricing for the broad range of markets considered in this paper is not available. As an alternative, a number of different variables are included which capture various elements of the current climate in the stock market. The most obvious measure of stock market performance is the return to a general indicator such as the market index. As discussed by Lowry (2003), market returns potentially capture changes in investment demand, investor sentiment as well as other unknown dynamics. As such, the total return to a market index is calculated based on the movement of an accumulated market index over the course of the year. (6) A second measure of stock market performance is dividend yield which will fall in a rising market, assuming constant dividends. The likelihood of capital gains may attract investors to a new listing in which case a negative relationship is expected, that is, falling yields signal rising prices and greater listing activity. (7) The dividend yield of a market is estimated as the total dividends distributed by domestic companies in the main index relative to the market capitalization. Other potential measures of market performance include the standard accounting ratios such as price-to-earnings and price-to-book value. Price-to-earnings is calculated by dividing the market capitalization of stocks included in the main index of the stock exchange by the total market earnings and price-to-book value is calculated in a similar fashion. Helwege and Liang (2004) argue listing activity is typically concentrated in particular segments of the stock market. Favourable expectations about a particular industry (proxied by the price-to-book ratio) are argued to lead to higher levels of listing activity. Abstracting this concept to a more macro level, the higher the price-to-book ratio of a country, the greater may be the level of aggregate listing activity.

Another variable which may capture the prevailing stock market conditions is the number of delistings. Information on delistings may capture the general sentiment of the overall market in a way that the previous variables do not, although the exact nature of the relationship is unclear. On the one hand, delistings may occur due to financial distress. In this case, an increase in either the volume or value of delistings will dampen investor enthusiasm for the market. On the other hand, delistings due to mergers, acquisitions, buy-outs etc. may prove beneficial to shareholders. In this case, an increase in either the volume or value of delistings will heighten investor enthusiasm for the market. Irrespective as to the cause, a significant relationship is hypothesised to exist between the number of delistings and investors willingness to participate in the stock market including purchasing equity offered via new listings. The exact nature of this relationship is an empirical issue. The variable included in the test equations is expressed relative to the total number of listings to account for any changes in the size of the market over the sample period.

In addition to these measures of stock market performance, a number of measures of stock market activity are also tested. These measures of activity capture the overall level of interest in the stock market, which is not necessarily reflected in the performance metrics (increased activity does not necessarily translate into rising prices). The measures include the number of transactions, the number of shares traded (both expressed per capita to account for growth in the economy), the total value of shares traded (expressed relative to market capitalisation to account for growth in the market), and the average trade value (the total value of share trading divided by the number of transactions). The greater the general level of investor interest in the stock market, as indicated by heightened values for each of these measures, the more potential interest investors may have in new listings.

A second factor which is thought to impact on listing activity is the business cycle. The general argument is that investment opportunities are linked to general business conditions in the economy. An improvement in the commercial climate leads to higher levels of investment and so, an increase in the number of firms attempting to false capital through new listings. As there is no direct measure of business conditions however, a proxy must be used such as a business climate index (Ljungqvist 1997) or an index of leading indicators (Brailsford et al. 2000). Unfortunately, such measures are not readily available for the broad cross section of countries considered in this study. As such, changing business conditions are proxied in this paper using the term premium which Chen, Roll and Ross (1986) argues captures future business conditions and Harvey (1988) suggests is a reasonable predictor of future economic growth. The terra premium is measured as the difference between the yield to maturity on ten year government bonds and the three month money market interest rate.

In addition to the term premium, a number of other general economic indicators are also tested including the inflation rate, the annual change in GDP and short term interest rates, proxied by the three month money market interest rate. Lowry (2003) hypothesises investor optimism is proxied by changes in GDP while Pagano, Panetta and Zingales (1998) proxy increases in sales with changes in GDP. Thus, while the previous literature provides a clear incentive to include GDP, its aggregate nature means that it is difficult to interpret the mechanism through which GDP impacts on listing activity. This is made even more difficult when the link to investment is considered, as listing may also be capturing changes in investment activity. Thus, a significant estimation coefficient for GDP must be interpreted with caution as it is may indicate a number of different influences on listing behaviour.

Following Benninga, Helmantel and Sarig (2003) and Lerner, Shane and Tsai (2003), a third factor which may affect listing activity is the performance of alternative investment vehicles such as bonds. The yield to these alternative investments may influence the value placed by the public on the cash flows of listed equity. An argument can be made that the lower the relative performance of these alternatives, the higher will be the publics demand for dividends and the greater the price they are prepared to pay to secure the rights to those cash flows. Thus, in periods of poor bond market performance, the greater may be the demand for equities and hence, the easier it will be for a company to issue shares. In this study, two general bond market indicators are considered, which capture both public and private sector bond market performance. The first is the change in the total value of domestic and foreign bond trading during the year. Domestic bonds include private and public sector listed bonds, governments and municipal bonds. Foreign bonds are bonds listed on the stock exchange and issued by non-resident institutions. The second general measure of bond market performance is the change in the market value of listed domestic and foreign bonds. Finally, the 10 year government bond yield is also included as a proxy measure of the long terra bond yield.

This annual dataset is sourced from the World Federation of Exchanges database except for the price-to-book ratio which is sourced from Standard and Poors Stock Market Factbook (2002), the GDP estimates which are taken from the World Bank World Development Indicators database and population data which is taken from the US Census Bureau, International Data Base (IDB). Where appropriate, all values are expressed in US dollars and are converted using yearend exchange rates. To give the reader a feeling for this dataset, the first four moments of the data are summarised in table 2 as well as the number of observations. Jarque Bera testing of the data revealed that all of the datasets are nonnormal at the 5% level. Prior to any empirical testing, it is important to consider the correlations of this dataset. Recall that the variables all capture information about the economy and it is possible that they may overlap to some extent. To this end, table 3 presents the correlation matrix for the data and generally the correlations are quite low (90 of 120 estimates are less than 0.1) and the average correlation across the dataset is only 0.034. The correlations of greatest importance are those which capture the relationship between variables which are included within a particular class of information. Thus, the correlation between the market return and the price-to-book value (0.465), the number of transactions and the number of shares traded (0.293) as well as the total value of shares traded (0.257), short terra interest rates and inflation (0.89), and the change in market value of bonds listed and the change in the total value of bond trading (0.346) are of particular concern. As a form of robustness testing, the analysis of section 4 is repeated with each variable considered individually and the tenor of the results (available on request) is unchanged.

4. The Determinants Of Listing Activity

Stationarity testing in the context of small panel data sets is a non-trivial issue which the literature is yet to fully resolve. (8) As such, formal stationarity testing cannot solely be relied upon and theoretical considerations must also come into play. With this in mind, the Im, Pesaran and Shin (2003) (IPS hereafter) stationarity test is applied to the panel data of the number of new listings and the critical value fails to reject the null of a unit root at the 5% level. This is consistent with expectations as Lowry (2003) argues that it is unlikely that the volume of new listings has a tendency to revert to some normal level of activity which suggests that the data may be nonstationary. To overcome this problem, a relative measure of listing activity is tested in this paper in which the total number of newly listed firms each year is expressed as a percentage of the total number of listed firms. The IPS test applied to the relative number of new listings data generates a p-value of 0.0361 and so rejects the unit root null.

A hot issue market signals an increased willingness of individuals to absorb equity issued by newly listed companies into their portfolios. The factors which influence this change in attitude may be a part of an overall trend in which investors have an increased demand for new equity per se. If this is the case, then both public and private firms may attempt to take advantage of this change in investor attitudes to raise capital. To test this proposition, a pooled cross sectional time series equation of the following form may be estimated:

[LIST.sub.i,t] = [[alpha].sub.0] + [[alpha].sub.1] [SEI.sub.i,t] + [[epsilon].sub.t] (1)

where [LIST.sub.i,t] is relative listing volume for exchange i in year t, SEI is a measure of the amount of capital raised by companies which already have shares issued (i.e. Seasoned Equity Issues) expressed as a percentage of market capitalisation, (9) [alpha] are parameters to be estimated and [epsilon] is the equation error terra which is modelled as an AR(l) process to account for serial correlation in the residuals. The results of this regression are presented in table 4. The estimated [[alpha].sub.1] coefficient suggests that there is a positive and significant contemporaneous relationship between the number of seasoned equity offerings and relative listing volume. (10) This evidence supports the hypothesis that in a hot issue market, the public has an increased appetite for both seasoned and unseasoned equity offerings. The AR(l) terra is highly significant (DW=2.27) and the estimated adjusted [R.sup.2] coefficient decreases from 0.517 to 0.049 where it is excluded from the regression equation. This suggests that a strong correlation exists between past and current listing volume which is consistent with the clustering characteristic observed in the previous literature. The remainder of this paper is concerned with identifying other factors beyond past listing volume, which may influence the willingness of investors to absorb equity from newly listed companies.

To test the determinants of listing activity, a number of explanatory variables related to stock and bond market performance as well as general business conditions in the economy may be considered. Following Lowry (2003), potential problems with multicolinearity are avoided by using a 'block' testing procedure in which the potential determinants of listing activity are grouped together and each group is considered separately. Once the significance of each proxy variable has been established within this group framework, a final regression equation is to be estimated that brings together each of these individually significant variables. This final equation will provide some insights into the collective explanatory power of these individual proxy variables.

As a first step in the analysis, relative listing volume may be regressed against a series of general indicators of stock market performance. The pooled equation to be tested takes the following form:

[LIST.sub.i,t] = [[alpha].sub.0] + [[alpha].sub.1] [DY.sub.i,t] + [[alpha].sub.2] + [PER.sub.i,t] + [[alpha].sub.3] + [PBV.sub.i,t] + [[alpha].sub.4] + [R.sup.Mkt.sub.i,t] + [[alpha].sub.5] + [Delist.sub.i,t] + [[epsilon].sub.t] (2)

where DY is the dividend yield, PER is the price to earnings ratio, PBV is the price-to-book value ratio, [R.sup.Mkt] is the return to a general stock market indicator and Delist is the number of companies delisted during the year relative to the total number of listed companies. The estimation results for this equation are presented as model (2) in table 4 and PB V is positive and significant indicating that higher firm values relative to their asset base result in higher levels of listing activity. This evidence reinforces the results of Pagano, Panetta and Zingales (1998) who found that companies with a higher market-to-book ratios are more likely to go public. The market return variable is also significant and the negative sign provides support for the arguments of Baker and Wurgler (2000) who argue that investors are overoptimistic and firms will time their sale of equity to coincide with periods of high stock prices. As investor optimism returns to more rational levels, stock market prices across the board will fall. Although dividend yield is insignificant, the negative sign provides some support for the hypothesis that in times of rising prices, and so falling dividend yields, greater listing activity may be observed. The adjusted [R.sup.2] of equation (2) in the absence of the AR(l) term is 0.068 which suggests that stock market performance indicators have only limited economic importance in determining listing activity.

To provide further insights into the nature of the relationship between market returns and listing volume, a dynamic version of equation (2) may be estimated and the results are presented as model (2a) in table 4. Where the model is re-estimated with a lagged market return coefficient, the estimate of [[alpha].sub.4] is positive and significant. In general, this evidence on market returns suggest that firms decide to list during the windows of opportunity created by a rising market and, as investor overconfidence is replaced with rationality, share prices fall to their long run values.

As an alternative to stock market performance, stock market activity may influence the decision to list. This may be tested using a pooled equation of the following form:

[LIST.sub.i,t] = [[alpha].sub.0] + [[alpha].sub.1] [NT.sub.i,t] + [[alpha].sub.2] [NST.sub.i,t] + [[alpha].sub.3] [ATV.sub.i,t] + [[alpha].sub.4] [TVST.sub.i,t] + [[epsilon].sub.t] (3)

where NT is the number of transactions per capita, NST is the number of shares traded per capita, A TV is the average trade value relative to total market capitalisation and TVST is the total value of shares traded relative to total market capitalisation. The estimation results for this equation are presented as model 3 in table 4 and the number of transactions per capita and the average value of shares traded are both positive and significant. To the extent that greater trading volume as well as increased average trade size reflect positive sentiment about the share market, these results reinforce earlier evidence of equation (2) that increased listing activity depends on recent conditions in the stock market. The ability of the number of transactions and the average trade value to explain listing activity is low however, as the adjusted [R.sub.2] for this model in the absence of the AR(1) terra is only 0.038. Where listing activity is simultaneously regressed against all nine measures of stock market performance and activity, the price-to-book value, market return and average trade value variable retain their significance and sign. The adjusted [R.sup.2] in the absence of the AR(l) terra is only 0.076 however, which represents only a slight improvement to equation (2).

Beyond stock market performance, business conditions may also impact on listing volume and to test this proposition, the following pooled equation is estimated:

[LIST.sub.i,t] = [[alpha].sub.0] + [[alpha].sub.1] [INF.sub.i,t] + [[alpha].sub.2] [GDP.sub.i,t] + [[alpha].sub.3] [STi.sub.i,t] + [[alpha].sub.4] [iSPREAD.sub.i,t] + [[epsilon].sub.t] (4)

where INF is the annual inflation estimate for the country in which the stock market resides, GDP is the change in annual gross domestic product, STi is the short terra interest rate and iSPREAD is the spread between short terra and long terra interest rates. The estimation output for this equation is presented as model 4 in table 4 and none of the variables are significant.

The final set of variables considered in this paper capture information relating to the bond market, which competes with shares for a finite pool of investment funds. Where expected bond market yields exceed the expected returns of the share market, investors may be hesitant to commit their funds to an new listing. To test this proposition, the following pooled equation may be estimated:

[LIST.sup.i,t] = [[alpha].sub.0] + [[alpha].sub.1] [MVB.sub.i,t] + [[alpha].sub.2] [VBT.sub.i,t] + [[alpha].sub.3] LTY + [[epsilon].sub.t] (5)

where MVB is the change in the market value of bonds, VBT is the change in the value of bond trading and LTY is a measure of the long term bond yield. The estimated output for this pooled regression equation is presented as model 5 in table 4 and none of the estimated coefficients are individually significant.

The estimation results for equations (2) to (5) suggest that stock market performance and activity, as proxied by price-to-book value ratio, market return, the number of transactions and average trade value, influence the level of listing activity. To determine the combined effect of these individually significant variables, a final parsimonious pooled regression equation is estimated which takes the following form:

[LIST.sub.i,t] = [[alpha].sub.0] + [[alpha].sub.1] [PVB.sub.i,t] + [[alpha].sub.2] [R.sup.Mkt.sub.i,t] + [[alpha].sub.3] [NT.sub.i,t] + [[alpha].sub.4][ATV.sub.i,t] + [[epsilon].sub.t] (6)

The estimation output for this model is presented as model 6 in table 4 and each of the estimated variables retain their sign and level of significance. The overall explanatory power of this equation is 0.536 and in the absence of the AR(l) term, this falls to 0.077. Thus, these stock market indicators together explain only 7% of listing volume and the most significant influence on the number of new listings is past listing activity.

As a form of robustness testing to the previous results and to alleviate any concerns as to nonstationarity of the dependent variable, the estimation results of this section are retested using the raw listing data. The results are presented in table 5 and may be summarised as follows. A positive and significant relationship is found between seasoned equity issues and listing activity. Thus, consistent with the earlier evidence, greater listing activity would seem to reflect a heightened appetite for investors for new equity per se. In terms of the ability of the explanatory variables to explain the different levels of listing activity across the exchanges considered in the sample, the results are broadly consistent to those already discussed. Specifically, of the stock market performance indicators, price-to-book value and market returns (including lagged market returns) are again found to be significant and their signs are consistent to the earlier set of results. Only the average trade value stock market activity indicator is significant as the number of transactions variable, although still positive is now insignificant. Of the general business conditions variables and the bond market indicators, none are significant. Thus, the parsimonious model presented as model (6), contains three significant variables each of which retains their sign and significance when compared to their block equation counterparts. Overall, the explanatory power of these equations is low and the most significant factor in determining current listing activity is again past listing activity. To the extent that the estimation results are largely insensitive to the specification of the dependent variable as raw listing data or relative listing data, the remainder of this paper shall focus on the relative version, which is theoretically desirable and is consistent with the previous literature (see Lowry 2003).

4.1 Developed and Emerging Market Analysis

The sample of stock markets considered in this study may be classified as belonging to a country which is either emerging or developed. This distinction may prove to be important as emerging markets have been found to exhibit a number of unique characteristics which distinguish them from their developed counterparts. For example, corporate structures in South East Asia are typically more focussed around the founding family and relationship banking means that commercial criteria are less relevant to securing bank capital. Further, emerging markets have lower levels of investor protection, a lack of comparable accounting standards, greater information asymmetries, and so on (see Bekaert & Harvey 2004). These differences mean that the market capitalisation, turnover and, most importantly, the number of listed companies in emerging stock markets is typically lower compared to other nonmarkets. In the context of the current paper, it is interesting to consider whether listing activity and the determinants of listing are different between developed and emerging markets. To test this proposition, the analysis of section 4 may be replicated distinguishing between stock markets in emerging and developed countries which are classified based on the Standard and Poors Stock Market Factbook, 2003. (11) The estimation results are presented in tables 6 and 7 for the developed and emerging markets, respectively. (12) For both the developed and emerging markets, the same positive relationship is found between listing activity and the capital raising efforts of already listed companies (results column 1). In terms of the determinants of listing activity in developed countries stock markets, the price-to-book value, (lagged) market return, number of transactions and average trade value are again significant and retain their estimated signs.

In addition to these four previously identified factors, a number of other variables are also significant in the context of the developed markets (table 6). Firstly, the number of delistings is significant and the negative sign signifies an increase in delistings is associated with a fall in the number of new listings. This result is consistent with the hypothesis that delistings signify bad news which dampens investor enthusiasm for the market and adversely impact the ability of firms to list. A second variable which is significant is the short term interest rate. To the extent that short terra interest rates typically respond to economic growth, this variable is argued to capture information as to the current business conditions in the economy. This suggests that heightened listing activity will follow a rise short terra interest rates rise (perhaps reflecting an improving economic outlook). A final model may be estimated which draws together each of these significant explanatory variables and the estimation result is presented in results column (6) of table 6. Each of the estimated variables retains their sign although the market return and short term interest rate are no longer significant. Comparing the adjusted R2 for this set of equations in table 6 to their whole sample counterparts, the explanatory power of equations (2) and (4) has increased with the addition of two more significant variables. The coefficient of determination for the combined significant factors regression equation is 17.0% compared to 7.7% across the whole sample. Thus, for these developed countries, stock market and business conditions appear to possess greater explanatory power in accounting for the number of new listings. The most significant determinant of current listing activity however, continues to be past listing activity as with the inclusion of an AR(l) term the adjusted [R.sup.2] increases to 57%.

Table 7 presents the equivalent analysis for the sample of emerging countries. The most interesting feature of the estimated results for this subsample of emerging stock markets is the general lack of significance of the variables tested. The (lagged) market return and short term interest rate variables are significant, although the estimated sign for the latter is opposite to that previously estimated. This explains why the short term interest rate coefficient is insignificant in the whole sample results as the positive coefficient estimated using the developed country data is offset by the negative coefficient estimated using the emerging market data. The interest rate spread is also significant in equation (4) of table 7, and the adjusted [R.sup.2] without the AR(I) term is 0.159. Where these three significant variables are jointly considered (equation 6), only the short term interest rate variable remains significant and the explanatory power of this equation is 33.5% or 7.6% in the absence of the AR(l) term. It would appear that the significance of the interest rate variables in this emerging market analysis is solely driven by the inclusion of the Turkish data (short term interest rates averaged 97.86% for the sample period). If Istanbul is excluded from the analysis of the developing countries, none of the estimated coefficients in equation (6) are significant and the adjusted [R.sup.2] is 0.232 (-0.037 excluding the AR(1) term).

In general, this analysis suggests that the lessons learned about the determinants of listing activity in the context of developed countries do not necessarily translate into knowledge about emerging markets. While listing activity in developed markets exhibits a clear link to stock market and business conditions, the same cannot be said for emerging markets. For both emerging and developed markets however, the most significant determinant of current listing activity appears to be past listing volume. Further, the analysis of emerging markets data highlights the need for careful analysis which takes into account the extreme values frequently observed in these countries.

4.2 Listing Activity and Investment

Mikkelson, Partch and Shah (1997) surveyed the prospectuses of issuing companies and found that 85% stated that the IPO was to provide working capital and 64% wanted to raise money for new investments. This would tend to suggest that the level of private business investment in the economy should increase following heightened periods of listing activity. Pagano, Panetta and Zingales (1998) however, found evidence to suggest that investment typically declines after a period of heightened activity. This is an interesting contradiction and it is possible to extend the analysis of this paper to provide some insights into the nature of the relationship between investment and listing. To this end, business sector investment data (measured in US dollars) is sourced from the OECD Economic Outlook database. This data was not available for South Africa, Luxembourg, Tel-Aviv and Warsaw, the four South American exchanges, and the Asian exchanges except for Korea and Japan.

Table 8 presents the estimation results of a model which estimates the impact of listing activity on investment which is expressed as a percentage of GDP. Where relative listing volume is considered, a positive contemporaneous relationship is generated which is significant although the [bar.R.sup.2] without the AR(l) term is only 0.021. Where a lagged version of this model is estimated, a positive result is again generated however, this coefficient is not significant. A model which includes both a contemporaneous and a lagged measure of listing activity generates a positive coefficient on the current period estimate but neither coefficients are significant and the [bar.R.sup.2] excluding the AR(1) term is 0.018. Where an alternative measure of listing activity is considered which focuses on the value of capital raised via listing, the contemporaneous model fails to produce a significant coefficient. A model with a lagged listing measure however, is negative and significant indicating a lower level of investment following heightened capital raising by private firms. Where both terms are simultaneously estimated, they are significant however, the [bar.R.sup.2] of this equation excluding the AR(l) term is negative. In general, these results provide some evidence of a relationship between listing activity and investment however, the explanatory power of these equations is low and in economic terms, it does not appear that listing activity has any influence on the level of investment.

Pagano Panetta and Zingales (1998) found that listing activity actually follows high investment and not vice versa. To test this proposition, the preceding analysis may be replicated where the relative measure of listing activity is the dependent variable and one and two period lagged investment is the independent variable. The results (not presented to conserve space) reveal that none of the estimated coefficients are significant. Thus, there is no evidence in the data set used in this paper to support the finding of Pagano, Panetta and Zingales (1998) that IPO activity follows high investment.

5. Conclusion

A literature exists which has considered the determinants of listing volume using data for a single country and a limited set of independent variables. In this study, the determinants of listing activity are considered, using a database of annual listing data for 38 international stock exchanges and a wide range of variables that capture stock and bond market information as well as general business conditions. In general, the results of this study provide some interesting new information on listing behaviour. Firstly, a positive contemporaneous relationship between listing activity and seasoned equity issues is found. This suggests that hot issue markets signal an increased appetite of investors for new equity per se. Thus, not only are companies able to list with relative ease during these periods, but companies which are already listed are also able to take advantage of the prevailing market conditions and raise capital.

A number of different variables are considered which capture information about conditions in the stock and bond market as well as general business conditions in the economy. Distinguishing between those exchanges which operate in developed markets and those in emerging countries, some interesting differences are uncovered. In terms of developed countries' stock markets, a range of variables were found to have a statistically significant impact on listing activity which capture stock market and business conditions. The ability of these factors to explain listing activity is limited, however, and the single most important factor in explaining the current level of listing activity is past volume. For the sample of stock markets in emerging countries, none of the independent variables tested possessed any real explanatory power. Further, the ability of past listing activity to explain current values was also less, compared to the developed market equivalent analysis. Previous surveys of company prospectuses suggest most IPOs are undertaken to raise working capital primarily for new investments. A study of the nature of the relationship between listing activity and investment undertaken in this paper finds no economically significant evidence of increased investment in the year of, or the year after, listing.

One limitation of the database used in this study relates to the aggregative nature of the information. The composition of firms choosing to list has changed over the years, most notably in terms of the recent trend toward technology stocks going public. The determinants of listing volume may well differ for different industry groupings and the current dataset is unable to provide any insights into this issue, and the author commends it as a further area of research. A further weakness of this study is the lack of information as to the level of underpricing, which has been found to be an important determinant of listing activity. Unfortunately, a time series of underpricing for the broad range of markets considered in this paper is not available and its inclusion would surely increase the explanatory power of the estimated equations.

The author would like to thank an anonymous referee. Part of the research for this paper was completed while the author was on sabbatical at the CERF, Cambridge University.

(Date of receipt of final transcript: December 4, 2006. Accepted by Garry Twite, Doug Foster & David Gallagher, Area Editors.)

References

Affleck-Graves, J., Hegde, S., Miller, R.E. & Reilly, F.K. 1993, 'The effect of the trading system on the underpricing of initial public offerings', Financial Management, vol. 22, pp. 99-108.

Baker, M. & Wurgler, J. 2000, 'The equity share in new issues and aggregate stock returns', Journal of Finance, vol. 55, pp. 2219-57.

Baltagi, B.H. & Kao, C. 2000, 'Nonstationary panels, cointegration in panels and dynamic panels: A survey' in Advances in Econometrics, ed. B.H. Baltagi, vol. 15, pp. 7-52.

Bekaert, G. & Harvey, C.R. eds. 2004, 'Emerging Markets' Elgar Reference Collection, International Library of Critical Writings in Financial Economics, vol. 14, Elgar, Cheltenham, U.K. and Northampton, Mass.

Benninga, S., Helmantel, M. & Sarig, O. 2003, 'The timing of initial public offerings', working paper, The Wharton School, University of Pennsylvania.

Blondell, D., Hoang, P., Powell, J.G. & Shi, J. 2002, 'Detection of financial time series turning points : A new CUSUM approach applied to IPO Cycles', Review of Quantitative Finance and Accounting, vol. 18, pp. 293-315.

Brailsford, T., Heaney, R., Powell, J. & Shi, J. 2000, 'Hot and cold IPO markets: Identification using a regime switching model', Multinational Finance Journal, vol. 4, pp. 35-68.

Chen, N.F., Roll, R. & Ross, S.A. 1986, 'Economic forces and the stock market', Journal of Business, vol. 56, pp. 383-403.

Choe, H., Masulis, R.W. & Nanda, V. 1993, 'Common stock offerings across the business cycle: Theory and evidence', Journal of Empirical Finance, vol. 3, pp. 3-31.

Cook, D.O., Jarrell, S.L. & Kieschnick, R. 2003, 'Investor sentiment and IPO cycles', working paper, University of Texas at Dallas.

Derrien, F. & Womack, K.L. 2003, 'Auctions vs. bookbuilding and the control of underpricing in hot IPO markets', Review of Financial Studies, vol. 16, pp. 31-61.

Harvey, C. 1988, 'The real term structure and consumption growth', Journal of Financial Economics, vol. 22, pp. 305-33.

Harvey, C. 1995, 'Predictable risk and returns in emerging markets', The Review of Financial Studies, vol. 8, pp. 773-816.

Helwege, J. & Liang, N. 2004, 'Initial public offerings in hot and cold markets', Journal of Financial and Quantitative Analysis, vol. 39, pp. 541-71.

Ibbotson, R.G. & Jaffe, J.F. 1975, 'Hot issue markets', Journal of Finance, vol. 30, pp. 1027-42.

Ibbotson, R.G., Sindelar, J.L. & Ritter, J.R. 1988, 'Initial public offerings', Journal of Applied Corporate Finance, vol. 1, pp. 37-45.

Ibbotson, R.G., Sindelar, J.L. & Ritter, J.R. 1994, 'The market's problems with the pricing of initial public offerings', Journal of Applied Corporate Finance, vol. 7, pp. 66-74.

Im, K.S., Pesaran, M.H. & Shin, Y. 2003, 'Testing for unit roots in heterogeneous panels', Journal of Econometrics, vol. 115, pp. 53-74.

Jegadeesh, N, Weinstein, M. & Welch, I. 1993, 'An empirical investigation of IPO returns and subsequent equity offerings', Journal of Financial Economics, vol. 34, pp. 153-75.

La Porta, R., Lopes-de-Silanes, F., Schleifer, A. & Vishny, R.W. 1997, 'Legal determinants of external finance', Journal of Finance, vol. 52, pp. 1131-50.

Lerner, J., Shane, H. & Tsai, A. 2003, 'do equity financing cycles matter? Evidence from biotechnology alliances', Journal of Financial Economics, vol. 67, pp. 411-46

Ljungqvist, A.P. 1997, 'Pricing of initial public offerings: Further evidence from Germany', European Economic Review, vol. 41, pp. 1309-20.

Loughran, T. & Ritter, J.R. 2002, 'Why don't issuers get upset about leaving money on the table in IPOs?', Review of Financial Studies, vol. 15, pp. 413-43.

Loughran, T., Ritter, J.R. & Rudqvist, K. 1994, 'Initial public offerings: International evidence', Pacific-Basin Finance Journal, vol. 2, pp. 165-99.

Lowry, M. 2003, 'Why does IPO volume fluctuate so much?', Journal of Financial Economics, vol. 67, pp. 3-40.

Lowry, M. & Schwert, G.W. 2002, 'Biases in the IPO pricing process', Journal of Finance, vol. 67, pp. 1171-98.

Lucas, D. & McDonald, R. 1990, 'Equity issues and stock price dynamics', Journal of Finance, vol. 45, pp. 1019-43.

Mikkelson, W.H., Partch, M.M. & Shah, K. 1997, 'Ownership and operating performance of companies that go public', Journal of Financial Economics, vol. 44, pp. 281-307.

Pagano, M., Panetta, F. & Zingales, L. 1998, 'Why companies go public? An empirical analysis', Journal of Finance, vol. 53, pp. 27-64.

Ritter, J.R. 1984, 'The 'hot issue' market of 1980', The Journal of Business, vol. 57, pp. 215-40.

Ritter, J.R. & Welch, I. 2002, 'A Review of IPO Activity, Pricing and Allocations', Journal of Finance, vol. 57, pp. 1795-828.

Rock, K. 1986, 'Why new issues are underpriced', Journal of Financial Economics, vol. 15, pp. 187-212.

Schultz, P. 2003, 'The timing of initial public offerings', working paper, University of Notre Dame.

Vitols, S. 2001, 'Frankfurt's neuer market and the IPO explosion: Is germany on the road to silicon valley?', Economy and Society, vol. 30, pp. 553-64.

Welch, I. 1989, 'Seasoned offerings, imitation costs, and the underpricing of initial public offerings', Journal of Finance, vol. 44, pp. 421-49.

Welch, I. 1996, 'Equity offerings following the IPO: Theory and evidence', Journal of Corporate Finance, vol. 2, pp. 227-59.

Michael D. McKenzi School of Economics and Finance, RMIT University, GPO Box 2476V, Melbourne, VIC 3001. Email: [e-mail]

Michael D. McKenzie ([dagger]), School of Economics and Finance, RMIT University, GPO Box 2476V, Melbourne, VIC 3001. Email: [e-mail]

The author would like to thank an anonymous referee. Part of the research for this paper was completed while the author was on sabbatical at the CERF, Cambridge University.

(1.) The decision to list and the decision to go public and list (an IPO) are technically distinct, but closely related areas of the literature. As the data in this paper relates to information on new listings (which includes IPOs), the analysis is framed in the context of new listings but will inevitably be guided by the IPO literature.

(2.) Some researchers have attempted to statistically model the time series properties of listing activity. For example, Blondell, Hoang, Powell, and Shi (2002) use turning point theory to characterise IPO markets. Brailsford et al. (2000) use a regime-switching model to identify hot issue markets for US IPO data.

(3.) During the sample period, the Amsterdam, Brussels, Paris and Lisbon Exchanges merged to form the Euronext exchange and the Barcelona, Bilbao, Madrid and Valencia exchanges joined to form the Spanish Exchanges. To construct the database used in this paper, all data for these individual exchanges was amalgamated prior to the merger. The data for the Osaka and Tokyo exchanges is also combined as some of the data is only available for Japan as a whole.

(4.) The value of capital raised by newly issued shares is ah alternative measure of listing activity. This may be an important distinction as stock exchanges have different listing rules regarding the minimum level of capitalisation. For example, during the dot.com boom in 2000, the NASDAQ had 7.8 listings for every 1 NYSE listing, but the total amount of capital raised by new listings on the NYSE was 85% higher. Ah analysis of the determinants of the amount of capital raised by newly issued shares is also undertaken in this paper and the unreported results add little to the discussion. They are available on request.

(5.) The evidence suggests that companies typically leave considerable 'money on the table' and estimates range from 256% for China 'A' shares to 5.4% in Denmark (based on an updated version of Loughran, Ritter & Rudqvist 1994, available at http://bear.cba.ufl.edu/ritter). These differences have been attributed to the institutional environment in which the IPO takes place (see Affleck-Graves, Hegde, Miller & Reilly 1993; Lowry & Schwert 2002; Derrien & Womack 2003). A related area of the literature has focussed on the post-listing performance of companies which have recently gone public. Research has found that IPOs tend to underperform in the long run relative to market and industry benchmarks although the evidence is sensitive to the econometric method, sample choice and benchmark employed. See Ritter and Welch (2002) for a survey of this literature.

(6.) For each exchange, the general market index used is--Amex: Amex Composite Index, Mexico: Price & Quotations Index (IPC), Nasdaq: Composite, NYSE: NYSE Composite Index, TSX Toronto Stock Exchange: TSE 300 Composite Index, Buenos Aires: General Index, Lima: General Index IGBVL, Santiago: IGPA, Sao Paulo: Bovespa, Athens: ASE Composite Price Index, Copenhagen: Total Share Index, Deutsche Borse: DAX Return, Euronext: average of CBS Ali Share Spot Return Index, PSI, General SBF 250 and X, Helsinki: HEX Ali Share Index, Irish: ISEQ Overall Index, Istanbul: ISE National 100, Italy: MIB Index, Johannesburg: JSE Actuaries Ali Share Index, London: FTSE 100, Luxembourg: Shares Price Index, Oslo: Oslo SE Total Index, Spanish Exchanges: average of BCN Global 100, Bolsa Bilbao 2000, IGBM and IGBV Index, Stockholmsborsen: SX General, Swiss Exchange: Swiss Performance Index (SPI), Tel-Aviv: General Share Index, Vienna: Wiener Borse Index, Warsaw: WIG, Australia: ASX/S&P Ali Ordinaries Index, Hong Kong: Ali Ordinaries Index, Jakarta: JSX Composite Index, Japan: Average of Osaka 300 and TOPIX, Korea: KOSPI, Kuala Lumpur: KLSE Composite Index, New Zealand: NZSE Gross Index, Philippine: PSE Composite Index, Singapore: Ali Sing Equities Index, Taiwan: TSE Weighted Stock (TAIEX), Thailand: SET.

(7.) It is possible to argue that where investors are attracted to value stocks, the opposite may be true. As the majority of IPOs do not pay dividends however, this is unlikely and so a negative relationship is hypothesised.

(8.) See Baltagi and Kao (2000) for a survey of the recent literature on stationarity testing and panel data.

(9.) The IPS tests applied to the raw data generated a p-value of 0.0006 while the data expressed as a percentage of market capitalisation generated a p-value of 0.0008. The reported estimation results in table 4 are not sensitive to the choice of independent variable.

(10.) Jegadeesh, Weinstein and Welch (1993) and Welch (1989, 1996) suggest that firms underprice their IPO to create a market for subsequent seasoned equity offerings. This suggests a heightened level of new listings may lead to increased future seasoned equity issues. Unfortunately, the current database does not allow sufficient observations to test this proposition.

(11.) In this paper, the IFC definition of ah emerging stock market is adopted such that any stock market in a developing economy is classified as emerging. The implication is that the stock markets potential is linked to the future economic growth. As such, the list of emerging markets includes Mexico, Buenos Aires, Lima, Santiago, Sao Paulo, Istanbul, JSE South Africa, Tel-Aviv, Warsaw, Jakarta, Korea, Kuala Lumpur, Philippines, Taiwan and Thailand.

(12.) An alternative testing procedure is to retest the database using an augmented equation in which the development status of a stock market is captured using a dummy variable. The unreported results are consistent with those generated using separate databases. For example, where the parsimonious regression equation for the developed markets (equation 6 from table 6) is reestimated using a dummy variable to capture the emerging market effect, the results are as expected given the emerging market results presented for equation 6 in table 7. Without extensive Wald testing however, it is not clear whether the dummy variable augmented variables retain their significance and so a separate approach to testing is maintained.

Table 1
Number of New Listings by Stock Exchange

The following table presents annual information on the number of
domestic companies newly listed on the main and parallel markets for
each stock exchange.

                           1995    1996    1997    1998

                                Panel A. North America

Amex                        74      92      91      92
Mexico                       1      14      18       3
Nasdaq                     413     598     573     437
NYSE                       138     219     210     162
TSX Toronto Stock Exch.     86     160     186     113
Average                    142.4   216.6   215.6   161.4

                                Panel B. South America

Buenos Aires                 7       7       2       3
Lima                        30      32      34      17
Santiago                    12      14      12       5
Sao Paulo                   24      26      19      57
Average                     18.3    19.8    16.8    20.5

                             Panel C. Europe, Africa and
                                   Middle East

Athens                      18      20      12      23
Copenhagen                  10       6       5      13
Deutsche Borse              20      20      35      67
Euronext (ABPL)             38      68     117     274
Helsinki                    10       3      12      12
Irish                        1       1       9       3
Istanbul                    30      25      29      20
Italian Exchange            12      15      13      21
JSE South Africa            27      27      53     100
London                     285     347     217     169
Luxembourg                   1       1       2       2
Oslo                        26      20      47      27
Spanish Exchanges (BBMV)    22      25      91     211
Stockholm                   16      17      50      32
Swiss Exchange               6       6      13      17
Tel-Aviv                    19       7      13      14
Vienna                       4       2       4       2
Warsaw                      21      18      62      57
Average                     31.4    34.9    43.6    59.1

                                     Panel D. Asia

Australian                  36      67      80      54
Hong Kong                   26      48      81      31
Jakarta                     21      16      30       6
Japan (Osaka + Tokyo)       59      97      76      67
Korea                       28      51      23       3
Kuala Lumpur                51      92      86      28
New Zealand                  3       7      10       5
Philippine                  16      13       6       3
Singapore                   17      16      33      23
Taiwan                      41      36      26      35
Thailand                    28      40       5       1
Average                     29.6    43.9    41.5    23.3

Average for all countries   44.1    59.8    62.8    58.1

                           1999    2000    2001    2002

                                Panel A. North America

Amex                        83     108      39      35
Mexico                       3       2       3       5
Nasdaq                     553     486     123     121
NYSE                       123      62      93     118
TSX Toronto Stock Exch.    140     114      81     102
Average                    180.4   154.4    67.8    76.2

                                Panel B. South America

Buenos Aires                 3       8       3       0
Lima                        17      12       6       1
Santiago                     4       4       3       1
Sao Paulo                   15      22      10      13
Average                      9.8    11.5     5.5     3.8

                             Panel C. Europe, Africa and
                                   Middle East

Athens                      37      52      21      15
Copenhagen                   6       7       4       0
Deutsche Borse             134     135      21       6
Euronext (ABPL)            241     231      36      15
Helsinki                    27      19       9       2
Irish                        3       3       1       0
Istanbul                     9      36       1       5
Italian Exchange            33      48      18      13
JSE South Africa            72      14      11       8
London                     161     366     236     193
Luxembourg                   2       5       1       2
Oslo                         9      16      12       5
Spanish Exchanges (BBMV)   547     558     677     506
Stockholm                   47      44      19       7
Swiss Exchange              17      23      14       5
Tel-Aviv                    14      41      15       8
Vienna                       4       7       6       3
Warsaw                      28      13       9       5
Average                     77.3    89.9    61.7    44.3

                                     Panel D. Asia

Australian                 112     163      72      86
Hong Kong                   37      89      88     117
Jakarta                      9      21      31      22
Japan (Osaka + Tokyo)       99     264     147     135
Korea                       16       5      16      25
Kuala Lumpur                21      38      20      56
New Zealand                  4      24      14      15
Philippine                   4       6       3       5
Singapore                   41      75      29      28
Taiwan                      33      80      69      85
Thailand                     0       2      10      24
Average                     34.2    69.7    45.4    54.4

Average for all countries   71.3    84.3    51.9    47.2

Note: Sourced from the World Federation of Stock Exchanges.

Table 2
Summary of Dataset

The following table presents a summary of the dataset used for
analysis in this paper as described in Section 3.

Variable                                  Mean   Std. Dev.   Skewness

Dividend Yield                           2.494       1.195      1.098
P / E Ratio                             27.457      57.982      9.970
Price-Book Value Ratio                   2.242       1.220      2.565
Market Return                            4.511      30.751      0.763
Delistings / Total No. Listings          0.058       0.044      0.756
No. Transactions per Capita              0.001       0.001      3.602
No. Shares Traded per Capita             0.013       0.050      6.349
Avg. Trade Value (USD 000's)            36.887      41.903      2.017
Tot Val Shrs Traded / Mkt Cap.           0.950       1.465      6.077
Inflation                                6.230      13.024      4.527
Change in GDP                            2.929      11.426     -1.015
Short Term Interest Rates               10.078      17.340      5.298
Interest Rate Spread                     0.506       5.621      2.993
Change in Market Value of Bonds (%)     11.963     102.411     12.760
Change in Value of Bond Trading (%)    139.728    1709.569     15.722
Long Term Bond Yield                    12.537      40.502     11.926

                                                    Number of
Variable                              Kurtosis   Observations

Dividend Yield                           4.828            290
P / E Ratio                            128.048            292
Price-Book Value Ratio                  13.878            302
Market Return                            6.625            304
Delistings / Total No. Listings          3.087            296
No. Transactions per Capita             17.264            278
No. Shares Traded per Capita            48.900            294
Avg. Trade Value (USD 000's)             7.032            278
Tot Val Shrs Traded / Mkt Cap.          49.126            304
Inflation                               25.250            296
Change in GDP                            9.188            304
Short Term Interest Rates               35.045            293
Interest Rate Spread                    52.176            247
Change in Market Value of Bonds (%)    182.922            235
Change in Value of Bond Trading (%)    250.615            258
Long Term Bond Yield                   165.358            248

Table 3
Dataset Correlation Estimates

                                      (1)      (2)      (3)      (4)

(1) Dividend Yield                    1
(2) P / E Ratio                      -0.154    1
(3) Price-Book Value Ratio           -0.326    0.144    1
(4) Market Return                    -0.247    0.053    0.465    1
(5) Delistings/ No. Listings          0.054    0.011    0.143    0.012
(6) No. Transactions per Capita       0.119    0.009    0.048   -0.122
(7) No. Shares Traded per Capita     -0.002    0.309    0.000   -0.036
(8) Avg. Trade Value                 -0.052    0.040    0.195    0.086
(9) Tot Val Shrs Traded / Mkt Cap.   -0.049    0.065    0.200   -0.122
(10) Inflation                       -0.066    0.172    0.203    0.278
(11) GDP                              0.004   -0.123    0.023    0.023
(12) Short Term Interest Rates        0.029    0.119    0.172    0.263
(13) Interest Rate Spread            -0.051    0.182   -0.020   -0.086
(14) Market Value of Bonds (%)       -0.029    0.014    0.068    0.047
(15) Value of Bond Trading (%)       -0.004   -0.015    0.002    0.017
(16) Long Term Bond Yield             0.067    0.072    0.050    0.102

                                      (5)      (6)      (7)      (8)

(1) Dividend Yield
(2) P / E Ratio
(3) Price-Book Value Ratio
(4) Market Return
(5) Delistings/ No. Listings          1
(6) No. Transactions per Capita      -0.181    1
(7) No. Shares Traded per Capita      0.129    0.293    1
(8) Avg. Trade Value                  0.266   -0.187   -0.135    1
(9) Tot Val Shrs Traded / Mkt Cap.   -0.263    0.257   -0.030    0.093
(10) Inflation                       -0.196   -0.133    0.219   -0.180
(11) GDP                              0.056   -0.007   -0.009    0.044
(12) Short Term Interest Rates       -0.185   -0.125    0.241   -0.198
(13) Interest Rate Spread             0.129    0.058    0.101    0.106
(14) Market Value of Bonds (%)        0.114   -0.010   -0.005   -0.013
(15) Value of Bond Trading (%)        0.014    0.006   -0.005   -0.015
(16) Long Term Bond Yield            -0.025   -0.086    0.173   -0.107

                                      (9)      (10)     (11)     (12)

(1) Dividend Yield
(2) P / E Ratio
(3) Price-Book Value Ratio
(4) Market Return
(5) Delistings/ No. Listings
(6) No. Transactions per Capita
(7) No. Shares Traded per Capita
(8) Avg. Trade Value
(9) Tot Val Shrs Traded / Mkt Cap.    1
(10) Inflation                       -0.002    1
(11) GDP                              0.020   -0.130    1
(12) Short Term Interest Rates        0.012    0.890    0.002    1
(13) Interest Rate Spread             0.006   -0.334    0.016   -0.341
(14) Market Value of Bonds (%)        0.082    0.002    0.090    0.002
(15) Value of Bond Trading (%)       -0.009   -0.025   -0.003   -0.024
(16) Long Term Bond Yield            -0.023    0.369    0.015    0.391

                                      (13)     (14)     (15)

(1) Dividend Yield
(2) P / E Ratio
(3) Price-Book Value Ratio
(4) Market Return
(5) Delistings/ No. Listings
(6) No. Transactions per Capita
(7) No. Shares Traded per Capita
(8) Avg. Trade Value
(9) Tot Val Shrs Traded / Mkt Cap.
(10) Inflation
(11) GDP
(12) Short Term Interest Rates
(13) Interest Rate Spread             1
(14) Market Value of Bonds (%)        0.019    1
(15) Value of Bond Trading (%)        0.003    0.346    1
(16) Long Term Bond Yield            -0.038   -0.003   -0.014

Table 4
Regression Results for Relative Listing Activity

The following table presents the regression results where relative
listing volume across the sample of stock markets is specified as the
dependent variable. The relationship between new listings and seasoned
equity offerings is tested in equation (1). The determinants of
listing activity are tested in equations (2)--(6) where a number of
independent variables representing stock and bond market conditions as
well as the performance of the economy are considered.

                            (1)           (2)          (2a)
                         Listed Co.      Stock         Stock
                          Capital       Market        Market
                          Raising     Performance   Performance

Constant                   0.0528       0.0325        0.0357
                          (5.45)       (2.03)        (2.07)
Seasoned Equity Issues     0.6388
                          (4.05)
Div. Yield                             -0.0015        0.0031
                                       (0.46)        (1.02)
PE Ratio                                0.0001        0.0001
                                       (0.65)        (0.33)
PBV Ratio                               0.0168        0.0092
                                       (4.83)        (3.10)
Market Return                          -0.0004
                                       (4.87)
Lagged Market Return                                  0.0004
                                                     (6.33)
Delistings / No.                       -0.062        -0.122
Listings                               (0.62)        (1.08)
No. Transactions per
capita
No. Shares Traded per
capita
Avg. Trade Value

Total Value Shares
Traded / Mkt Cap.
Inflation

GDP

Short Term Interest
Rates
Interest Rate Spread

Change in Market
Value of Bonds
Change in Value of
Bond Trading
Long Term Bond Yield

AR(1)                      0.7007       0.6893        0.6943
                         (14.98)      (15.21)       (13.74)
D.W.                       2.274        2.319         1.936
Adj. [R.sup.2]             0.517        0.525         0.552
Adj. [R.sup.2] without     0.049        0.068         0.120
  AR(1)
SBC                       -2.337       -2.803        -2.837
No Obs                   224          236           202

                             (3)           (4)
                            Stock       Business
                           Market      Conditions
                          Activity

Constant                   0.0318        0.0665
                          (2.97)        (6.45)
Seasoned Equity Issues

Div. Yield

PE Ratio

PBV Ratio

Market Return

Lagged Market Return

Delistings / No.
Listings
No. Transactions per      11.415
capita                    (2.45)
No. Shares Traded per     -0.0151
capita                    (0.17)
Avg. Trade Value           0.0005
                          (4.05)
Total Value Shares         0.0021
Traded / Mkt Cap.         (0.69)
Inflation                               -0.0006
                                        (1.11)
GDP                                      0.0001
                                        (0.25)
Short Term Interest                     -0.0007
Rates                                   (1.45)
Interest Rate Spread                    -0.0005
                                        (1.37)
Change in Market
Value of Bonds
Change in Value of
Bond Trading
Long Term Bond Yield

AR(1)                      0.6733        0.7218
                         (15.10)       (14.48)
D.W.                       2.254         2.142
Adj. [R.sup.2]             0.513         0.459
Adj. [R.sup.2] without     0.038        -0.004
  AR(1)
SBC                       -2.481        -2.340
No Obs                   271           203

                             (5)           (6)
                            Bond        Combined
                           Market      Significant
                         Performance     Factors

Constant                   0.0547        0.0107
                          (4.77)        (0.89)
Seasoned Equity Issues

Div. Yield

PE Ratio

PBV Ratio                                0.0138
                                        (4.04)
Market Return                           -0.0003
                                        (3.95)
Lagged Market Return

Delistings / No.
Listings
No. Transactions per                     8.6366
capita                                  (2.03)
No. Shares Traded per
capita
Avg. Trade Value                         0.0003
                                        (2.53)
Total Value Shares
Traded / Mkt Cap.
Inflation

GDP

Short Term Interest
Rates
Interest Rate Spread

Change in Market           0.0001
Value of Bonds            (0.03)
Change in Value of         0.0001
Bond Trading              (0.67)
Long Term Bond Yield       0.0001
                          (0.43)
AR(1)                      0.7468        0.6846
                         (11.91)       (15.47)
D.W.                       1.796         2.342
Adj. [R.sup.2]             0.492         0.536
Adj. [R.sup.2] without    -0.012         0.077
  AR(1)
SBC                       -2.346        -2.793
No Obs                   144           240

Table 5
Regression Results for Total Listing Activity

The following table presents the regression results where total listing
activity is specified as the dependent variable. The relationship
between new listings and seasoned equity offerings is tested in
equation (1). The determinants of listing activity are tested in
equations (2)--(6) where a number of independent variables representing
stock and bond market conditions as well as the performance of the
economy are considered.

                                  (1)           (2)          (2a)
                              Listed Co.       Stock         Stock
                                Capital       Market        Market
                                Raising     Performance   Performance

Constant                        50.8313       72.9979       63.4196
                                (1.64)        (1.63)        (1.52)

Seasoned Equity Issues           0.0027
                                (5.777)

Div. Yield                                    -3.1995        0.2348
                                              (0.94)        (0.06)

PE Ratio                                       0.0232        0.0386
                                              (0.65)        (0.99)

PBV Ratio                                      8.0555        3.1964
                                              (2.19)        (0.98)

Market Return                                 -0.2767
                                              (2.91)

Lagged Market Return                                         0.2123
                                                            (2.92)

Delistings / No. Listings                    -77.6258     -132.1343
                                              (0.72)        (1.04)

No. Transactions per
capita

No. Shares Traded per
capita

Avg. Trade Value

Total Value Shares
Traded / Mkt Cap.

Inflation

GDP

Short Term Interest
Rates

Interest Rate Spread

Value of Bonds

Change in Value of
Bond Trading

Long Term Bond Yield

AR(1)                            0.8939        0.9282        0.9131
                               (26.90)       (27.95)       (25.68)

D.W.                             2.28          2.18          2.10

Adj. [R.sup.2]                   0.8296        0.7739        0.7680

Adj. [R.sup.2] without           0.3511        0.0199        0.0208
AR(1)

SBC                             12.385        11.855        11.976

No. Observations               224           236           202

                                  (3)           (4)           (5)
                                 Stock       Business        Bond
                                Market      Conditions      Market
                               Activity                   Performance

Constant                        33.2769       74.9973     -538.1297
                                (1.03)        (1.90)        (0.25)

Seasoned Equity Issues

Div. Yield

PE Ratio

PBV Ratio

Market Return

Lagged Market Return

Delistings / No. Listings

No. Transactions per         11386.27
capita                          (1.62)

No. Shares Traded per           -6.8684
capita                          (0.05)

Avg. Trade Value                 1.5171
                                (8.12)

Total Value Shares               1.2085
Traded / Mkt Cap.               (0.28)

Inflation                                     -0.1367
                                              (0.17)

GDP                                            0.1020
                                              (0.27)

Short Term Interest                           -0.0602
Rates                                         (0.07)

Interest Rate Spread                          -0.2402
                                              (0.45)

Change in Market                                             0.0343
Value of Bonds                                              (1.36)

Change in Value of                                           0.0046
Bond Trading                                                (0.27)

Long Term Bond Yield                                        -0.0027
                                                            (0.04)

AR(1)                            0.8930        0.8981        1.0105
                               (28.17)       (27.29)       (28.14)

D.W.                             1.95          2.13          2.15

Adj. [R.sup.2]                   0.8345        0.7873        0.8487

Adj. [R.sup.2] without           0.2427       -0.0017        0.0506
AR(1)

SBC                             12.052        12.688        12.802

No. Observations               271           203           144

                                  (6)
                               Combined
                              Significant
                                Factors

Constant                        36.0001
                                (1.07)

Seasoned Equity Issues

Div. Yield

PE Ratio

PBV Ratio                        7.6102
                                (1.96)

Market Return                   -0.2103
                                (2.19)

Lagged Market Return

Delistings / No. Listings

No. Transactions per
capita

No. Shares Traded per
capita

Avg. Trade Value                21.2036
                                (6.95)

Total Value Shares
Traded / Mkt Cap.

Inflation

GDP

Short Term Interest
Rates

Interest Rate Spread

Change in Market
Value of Bonds

Change in Value of
Bond Trading

Long Term Bond Yield

AR(1)                            0.8997
                               (28.87)

D.W.                             2.07

Adj. [R.sup.2]                   0.8285

Adj. [R.sup.2] without           0.2053
AR(1)

SBC                             12.067

No. Observations               240

Table 6
Relative Listing Activity for Developed Countries

The following table presents the regression results where the
relatively listing activity in developed markets is specified as the
dependent variable. The relationship between seasoned and unseasoned
equity offerings is tested in equation (1). The determinants of listing
activity are tested in equations (2)--(6) where a member of independent
variables representing stock and bond market conditions as well as the
performance of the economy are considered.

                                   (1)           (2)          (2a)
                               Listed Co.       Stock         Stock
                                 Capital       Market        Market
                                 Raising     Performance   Performance

Constant                          0.0681        0.0509        0.0783
                                 (6.37)        (1.89)        (2.86)

Seasoned Equity                   0.4315
Issues                           (2.31)

Div. Yield                                     -0.0037       -0.0013
                                               (0.61)        (0.22)

PE Ratio                                        0.0001        0.0001
                                               (0.63)        (0.36)

PBV Ratio                                       0.0206        0.0093
                                               (4.83)        (2.040)

Market Return                                  -0.0004
                                               (2.76)

Lagged Market                                                 0.0006
Return                                                       (5.02)

Delistings / No.                               -0.3204       -0.3618
Listings                                       (2.26)       (10.40)

No. Transactions
per capita

No. Shares Traded
per capita

Avg. Trade Value

Total Value Shares
Traded / Mkt Cap.

Inflation

GDP

Short Term Interest
Rates

Interest Rate Spread

Change in Market
Value of Bonds

Change in Value of
Bond Trading

Long Term Bond
Yield

AR(1)                             0.6647        0.722         0.6856
                                (10.27)       (11.81)       (10.40)

D.W.                              2.196         2.356         2.11

Adj. [R.sup.2]                    0.446         0.512         0.5799

Adj. [R.sup.2] without            0.020         0.067         0.163
AR(1)

SBC                              -3.02         -2.974        -3.029

No. Observations                135           134           114

                                   (3)           (4)           (5)
                                  Stock       Business        Bond
                                 Market      Conditions      Market
                                Activity                   Performance

Constant                          0.0273        0.0296        0.0546
                                 (1.82)        (1.32)        (2.02)

Seasoned Equity
Issues

Div. Yield

PE Ratio

PBV Ratio

Market Return

Lagged Market
Return

Delistings / No.
Listings

No. Transactions                 25.3696
per capita                       (2.19)

No. Shares Traded                -0.0832
per capita                       (0.40)

Avg. Trade Value                  0.0006
                                 (4.10)

Total Value Shares                0.0001
Traded / Mkt Cap.                (0.04)

Inflation                                       0.0011
                                               (0.37)

GDP                                            -0.0005
                                               (1.19)

Short Term Interest                             0.0098
Rates                                          (2.06)

Interest Rate Spread                            0.0037
                                               (0.76)

Change in Market                                              0.0001
Value of Bonds                                               (0.64)

Change in Value of                                            0.0001
Bond Trading                                                 (0.66)

Long Term Bond                                                0.0056
Yield                                                        (1.15)

AR(1)                             0.6835        0.6529        0.7827
                                (10.46)       (10.64)       (10.07)

D.W.                              2.136         1.973         1.643

Adj. [R.sup.2]                    0.490         0.473         0.557

Adj. [R.sup.2] without            0.097         0.026         0.011
AR(1)

SBC                              -3.026        -3.567        -2.972

No. Observations                135           138           86

                                   (6)
                                Combined
                               Significant
                                 Factors

Constant                          0.0195
                                 (1.09)

Seasoned Equity
Issues

Div. Yield

PE Ratio

PBV Ratio                         0.0143
                                 (3.70)

Market Return                     0.0001
                                 (0.74)

Lagged Market
Return

Delistings / No.                 -0.2785
Listings                         (2.35)

No. Transactions                 13.1412
per capita                       (1.93)

No. Shares Traded
per capita

Avg. Trade Value                  0.0002
                                 (2.32)

Total Value Shares
Traded / Mkt Cap.

Inflation

GDP

Short Term Interest               0.0041
Rates                            (1.56)

Interest Rate Spread

Change in Market
Value of Bonds

Change in Value of
Bond Trading

Long Term Bond
Yield

AR(1)                             0.6948
                                (11.24)

D.W.                              2.275

Adj. [R.sup.2]                    0.570

Adj. [R.sup.2] without            0.170
AR(1)

SBC                              -3.062

No. Observations                139

Table 7
Relative Listing Activity for Emerging Markets

The following table presents the regression results where the total
number of listings in emerging markets is specified as the dependent
variable. The relationship between seasoned and unseasoned equity
offerings is tested in equation (1). The determinants of listing
activity are tested in equations (2)--(6) where a number of independent
variables representing stock and bond market conditions as well as the
performance of the economy are considered.

                                  (1)           (2)          (2a)
                              Listed Co.       Stock         Stock
                                Capital       Market        Market
                                Raising     Performance   Performance

Constant                         0.0236        0.0292        0.0160
                                (1.12)        (1.29)        (0.62)

Seasoned Equity                  1.1206
Issues                          (3.90)

Div. Yield                                    -0.0004        0.0030
                                              (0.10)        (0.79)

PE Ratio                                       0.0001        0.0001
                                              (0.67)        (0.02)

PBV Ratio                                      0.0045       -0.0005
                                              (0.70)        (0.10)

Market Return                                 -0.0002
                                              (2.13)

Lagged Market                                                0.0002
Return                                                      (2.96)

Delistings / No.                               0.1160        0.1144
Listings                                      (0.82)        (0.70)

No. Transactions
per capita

No. Shares Traded
per capita

Avg. Trade Value

Total Value
Shares Traded /
Mkt Cap.

Inflation

GDP

Short Term
Interest Rates

Interest Rate
Spread

Change in Market
Value of Bonds

Change in Value
of Bond Trading

Long Term Bond
Yield

AR(1)                            0.7573        0.6994        0.7113
                               (10.84)       (11.18)        (9.96)

D.W.                             2.166         2.327         1.76

Adj. [R.sup.2]                   0.599         0.567         0.535

Adj. [R.sup.2] without           0.095         0.029         0.044
AR(1)

SBC                             -2.572        -2.494        -2.577

No. Observations                84           102            88

                                  (3)           (4)           (5)
                                 Stock       Business        Bond
                                Market      Conditions      Market
                               Activity                   Performance

Constant                         0.0330        0.0489        0.0330
                                (1.82)        (2.15)        (3.41)

Seasoned Equity
Issues

Div. Yield

PE Ratio

PBV Ratio

Market Return

Lagged Market
Return

Delistings / No.
Listings

No. Transactions                 1.4998
per capita                      (0.22)

No. Shares Traded               -0.1241
per capita                      (1.05)

Avg. Trade Value                 0.0003
                                (0.51)

Total Value                      0.0114
Shares Traded /                 (1.53)
Mkt Cap.

Inflation                                     -0.0004
                                              (0.87)

GDP                                            0.0003
                                              (1.02)

Short Term                                    -0.0011
Interest Rates                                (2.43)

Interest Rate                                 -0.0006
Spread                                        (2.01)

Change in Market                                             0.0001
Value of Bonds                                              (0.61)

Change in Value                                              0.0001
of Bond Trading                                             (0.78)

Long Term Bond                                               0.0001
Yield                                                       (0.30)

AR(1)                            0.6758        0.8049        0.5191
                               (10.15)       (12.61)        (4.50)

D.W.                             2.379         1.934         1.927

Adj. [R.sup.2]                   0.522         0.357         0.232

Adj. [R.sup.2] without           0.006         0.159        -0.031
AR(1)

SBC                             -2.438        -3.567        -3.408

No. Observations                94            58            54

                                  (6)
                               Combined
                              Significant
                                Factors

Constant                         0.0466
                                (2.19)

Seasoned Equity
Issues

Div. Yield

PE Ratio

PBV Ratio

Market Return                    0.0001
                                (1.50)

Lagged Market
Return

Delistings / No.
Listings

No. Transactions
per capita

No. Shares Traded
per capita

Avg. Trade Value

Total Value
Shares Traded /
Mkt Cap.

Inflation

GDP

Short Term                      -0.0013
Interest Rates                  (3.36)

Interest Rate                   -0.0004
Spread                          (1.51)

Change in Market
Value of Bonds

Change in Value
of Bond Trading

Long Term Bond
Yield

AR(1)                            0.7983
                               (13.36)

D.W.                             1.961

Adj. [R.sup.2]                   0.335

Adj. [R.sup.2] without           0.076
AR(1)

SBC                             -3.539

No. Observations                62

Table 8
Regression Results for the Relationship
Between Investment and Listing Activity

Dependent Variable: Private Sector Investment as a Percentage of GDP

Constant                           0.1605      0.1637      0.1594
                                  (2.30)      (3.30)      13.21)
Listing Volume                     0.0485                  0.0711
                                  (2.08)                  (1.24)
Listing Volume (-1)                            0.0305     -0.018
                                              (1.63)      (0.41)
Listing Capital Raised

Listing Capital Raised (-1)

AR (1)                             1.0151      1.0234      1.0234
                                (191.70)    (177.05)    (177.55)
Adj. [R.sup.2]                     0.996       0.996       0.996
Adj. [R.sup.2] without AR (1)      0.021       0.015       0.018
D.W.                               2.058       1.880       1.886

Constant                           0.1790      0.1920      0.1833
                                  (2.23)      (3.57)      (3.50)
Listing Volume

Listing Volume (1)

Listing Capital Raised            -0.0066                  0.2229
                                  (0.10)                  (2.60)

Listing Capital Raised (1)                    -0.1044     -0.1935
                                              (2.84)      (3.91)

AR (1)                             1.0154      1.0244      1.0245
                                (173.97)    (167.01)    (171.11)
Adj. [R.sup.2]                     0.995       0.996       0.996
Adj. [R.sup.2] without AR (1)     -0.006      -0.006      -0.013
D.W.                               2.137       1.935       1.892