Macro-economic determinants of the IPO waves: An assessment of G-7 countries

Literature review

The majority of the studies in the literature have analyzed the trend of IPOs in terms of number and volume and have created the concepts known as IPO Wave and Hot and Cold Issue markets in the literature. In most of the studies in the literature, the markets in which the number and volume of IPOs increase gradually are defined as “hot issue markets”, while the markets where the number and volume of IPOs decrease are defined as “cold issue markets”. In hot issue markets, many companies can plan their IPOs for this period, as the investor’s demand for the stock market is increasing. This way, the company makes a successful public offering and the investors have the opportunity to obtain abnormal returns in the short term thanks to the high discounts they obtain.

Pástor and Veronesh (2005) developed an optimal IPO timing model for firms. They emphasized that many studies in the literature explained the change in IPO volume over time with the concept of market inefficiency, where IPOs are made when stocks are overvalued. Contrary to previous literature, they created a model in which the fluctuation in the volume of initial public offerings occurs without any mispricing mechanism and the volume of the IPO is more closely related to recent changes in stock prices rather than the level of market stock prices. In the model, the variables that companies take into account during the public offering are, respectively: expected market return, expected aggregate profitability and prior uncertainty about the average future profitability of IPOs. According to the results of the analysis, they determined that there were high market returns before the IPO waves and low market returns after the IPO waves.

Buttimer, Hyland and Sanders (2005), in their study, comparatively examined the waves of IPOs in the real estate investment trust (REIT) sector and the waves of general initial public offerings. In terms of initial returns, it was observed that the REIT sector provided lower returns than the general market average after the IPO waves. In addition, the long- term underperformance anomaly, which is frequently observed in stock markets according to many studies such as Ritter (1991), Loughran and Ritter (1997), Levis (1993), Schultz (2003) among others, has not been observed in the REIT sector. Based on the findings of their study, authors stated that the Capital Demand Hypothesis best describes the REIT public offering market.

Yung, Colak and Wang (2008) created a model in which real investment opportunities that change over time cause adverse selection in the market for IPOs. As a result of the analysis, it has been determined that economic expansions are highly correlated with the number of companies that go public, which is positively correlated with underpricing. In line with the model they created, greater cross-sectional return variance and higher delisting rates were detected for hot market public offerings.

Christoffersen, Nain and Tang (2010) examined the waves and the quality of IPOs. They stated that the quality of the public offering will not be the same in the first and later periods of the IPO wave. They noted that in the early stages of the IPO wave, the average quality of IPO stock is low when initial returns and IPO demand are low. However, later in the IPO wave, when initial returns and IPO demand are low, IPOs have better operating performance, higher market share, and consequently higher long-term abnormal returns. According to the results of the study, they found that institutional investors benefit from short-term high returns in early public offerings, but they prefer IPOs that are done in later stages of the IPO wave thanks to their higher quality in the long run.

Tran and Jeon (2011) analyzed the macroeconomic factors that were effective on the initial public offerings in the US capital market during the 1970-2005 period. By applying time series econometric methods in the analysis, they determined the long-term balance between macroeconomic variables and public offerings. As in many studies, Tran and Jeon also found that stock market return performance and volatility have the most significant impact on the timing of the IPO. In their study, they also examined the effects of interest rates and bond market returns. They found that there are not only long-term but also short-term statistically significant relationships between macroeconomic variables and IPO activities.

Angelini and Foglia (2018) examined the short- and long-term relationships between initial public offerings and macroeconomic factors in the UK stock market for the period 1996-2016. In the analysis, they sought answers to the questions of how macroeconomic conditions affect the initial public offerings and how long the recent shock effect lasted. Business cycle, volatility, interest rates and stock returns were used as macroeconomic variables. According to the results of the correlation analysis made in the study, it was determined that the business cycle, volatility and interest rate variables could explain the change in the number of IPOs. However, unlike many studies in the literature, no significant effect of stock market performance on the number and timing of IPOs could be detected.

Thanh (2020) used macro econometric models to examine the cycles of initial public offerings in US stock markets. According to the results of the time series analysis they applied in their studies, they determined the strong and negative effect of macroeconomic uncertainty on the public offering activity. It has been determined that a one-unit standard deviation increase in macroeconomic uncertainty reduces the monthly number of initial public offerings by approxi- mately four units in the long run. In the analysis, they also found that both the decrease in the number of IPO applications and the increase in withdrawn IPOs contributed to the decrease in the number of IPOs in response to an uncertainty shock.

Carosi and Mengoli (2021) analyzed local IPO waves in their studies. They analyzed public offerings on the basis of region and sector. They found that the waves of public offerings overlapped on a sectoral and regional basis. In other words, public offerings on a sector basis are similar to public offerings on a regional basis. They observed that IPOs at the beginning of the IPO wave were equally priced lower than IPOs at the end of the wave. According to the results of the analysis, it was determined that the IPO decision is sensitive not only to the high valuations of the companies in the same sector, but also to the high valuations of the companies in the same region but in different sectors.

The research conducted by Cerpentier et al. (2022) focuses on the impact of market timing on the capital structure of private firms raising initial equity crowdfunding (ECF). Unlike prior finance studies that primarily concentrate on publicly traded or IPO firms, this study uniquely delves into the capital structure dynamics of privately held entities. Analyzing firms funded through the two major UK ECF platforms, Crowdcube or Seedrs, the study differentiates between ECF campaigns launched in hot markets, characterized by high ECF volumes, and those in cold markets. Results indicate that firms in hot markets set higher targets, collect more overfunding, and consequently raise more equity capital compared to firms in cold markets. Surprisingly, there are no discernible differences in leverage ratios between firms in hot and cold markets during the year of the ECF campaign, contradicting the expectations derived from market timing theories of capital structure. This unexpected finding is attributed to hot-market ECF firms contemporaneously rebalancing their capital structure by attracting more debt, particularly financial debt.

Mehmood et al. (2021) comprehensively examines the phenomenon of underpricing in Initial Public Offerings (IPOs) across developed, developing, and emerging markets. Through an extensive analysis of existing literature, the review highlights a prevalent trend of higher underpricing in emerging markets compared to developing and developed markets. It identifies this discrepancy as emerging market issuers interpreting underpricing as a signal of quality IPOs, particularly where information asymmetry plays a significant role. Additionally, the review identifies various country-specific factors influencing the scale of underpricing across different markets, providing insight into potential primary determinants for future research.

Cao et al. (2021) investigate how market frictions affect managerial incentives and the organizational structure of new hedge funds. Their study develops a model wherein new managers actively seek accredited investors and exhibit stronger incentives to acquire managerial skills when faced with low investor demand. Empirical analysis using a merged database indicates that cold fund inceptions, identified ex-ante with low investor demand, outperform existing hedge funds, while hot inceptions facing high demand do not perform as well. This sheds light on the competitive advantages of cold stand-alone inceptions over various types of family-affiliated inceptions.

Dataset and sample structure

In the study, we examined whether macroeconomic variables have an effect on the frequency of public offerings. Since the aim is to determine the factors affecting the IPO wave through developed capital markets, the sample consists of G-7 countries. The analysis period covered the period 1999-2020, and annual data were used in the analysis.

Macroeconomic indicators such as annual consumer inflation, stock market returns and economic growth over the years (GDP growth) were used as independent variables. Explanations regarding on variables included in analysis are provided in Table 1 as follows.

Our preliminary expectations were a negative relationship between consumer inflation and the frequency of public offerings, and a positive relationship between stock returns and economic growth and frequency of public offerings.

Methodology

In the study, the relationship between the IPO waves (frequency) and macroeconomic variables were analyzed in terms of country sections. For this reason, panel data analysis method was used. In the analysis, it was deemed appropriate to apply the random effect model as the econometric model. While macroeconomic variables were included in the analysis as independent variables, the number of IPOs was included in the analysis as a dependent variable.

Since there was a cross-section dependency between the series in the analysis, the stationarity of the data was measured with the Peseran (2007) unit root test, which is one of the second-generation unit root tests. Afterwards, the Hausman test was applied to determine the econometric model to be applied. According to the results of the test, the application of the random effect model was deemed appropriate and the Driscoll-Kraay resistant standard estimator developed against deviations was used in the application of the model.

Descriptive statistics

Descriptive statistics for the sample used in the study are shown in Table 2 below.

Correlation analysis is one of the most typical analyzes used to examine the relationship between variables. Pearson’s correlation analysis was used in this study, as the data showed normal distribution.

The correlation relationship between the variables is shown in Table 3 below.

According to the results of the correlation test, since the degree of correlation between the variables varies between 0.05 and 0.49, there was a positive but weak relationship. The degree of correlation differed according to the variables used. For example, there was a very weak correlation of 0.05 between stock returns and inflation, while a weak correlation of 0.49 between the GDP growth and stock returns. A strong and very strong relationship between the variables in the correlation analysis should have a degree of correlation between 0.70-0.89 and 0.90-1.00, respectively. As a result, according to the results of the analysis, there is a weak relationship between the variables in this sample.

Unit root test

In order not to encounter the spurious regression problem in the analysis, it is necessary to measure the stationarity of the series before the regression analysis. However, in order to determine which stationarity test will be used, a cross-section dependency test should be applied. According to the result of the cross-sectional dependence, the appropriate unit root test was applied. Cross-section dependence was tested with Pesaran (2004) cross-section dependence method.

The hypotheses of this test were as follows:

H0: There is no cross-section dependence between the series.

H1: There is a cross-section dependence between the series.

The results of the test are shown in Table 4 below:

According to the results of the test, there was a cross-sectional dependence between the series. As a consequence, Pesaran (2007) second generation unit root test was applied, taking this stiuation into account.

The hypotheses of these tests were as follows:

H0: Series are not stationary.

H1: Series are stationary.

The results of the test are shown in Table 5 below.

According to the unit root test results, IPOs and Inflation were found to be stationary after taking the first difference, while the GDP growth and stock returns variables were found to be stationary at level values. Therefore, non-stationary IPOs and inflation variables were included in the regression analysis after taking their first difference. For this reason, a cointegration test was not be applied in the study.

Econometric model

A Hausman test was used to determine the econometric model to be established. The Hausman test results are summarized in Table 6 below.

According to this result, the random effect model was used as the econometric model.

Tests related to autocorrelation and heteroscedasticity problems in the model were performed in Table 7 and Table 8 below.

In the model, macroeconomic variables were used as independent variables, and IPO numbers were used as dependent variables, and a random effect regression analysis with the Driscoll-Kraay resistant standard estimator was applied.

Regression analysis results are shown in Table 9 as follows.

According to the results of the regression analysis, stock returns is the only variable that has a positive and significant effect on the number of IPOs. The effects of all other variables on the frequency of IPOs were not found significant. Based on the analysis findings, it can be stated that companies take into account the stock returns variable in timing the public offering. The correlation between stock returns variable and IPOs was found to be positive. In other words, as the stock market index returns increase, companies may choose to go public in order to enter the hot public offering market and maximize their issuance proceed (or reduce the underpricing degree of public offerings). The findings are compatible with many studies in the previous literature, especially Pástor and Veronesh (2005), and Tran and Jeon (2011).

Research limitations and further research suggestions

The study was limited to a sample comprising solely G-7 countries, representing developed markets. Consequently, the study lacks the capacity for comparative analysis and interpretation across different country groups, given the absence of results from emerging or underdeveloped markets. To attain more comprehensive and comparative outcomes, it is advised to replicate similar analyses within other country cohorts. Furthermore, the study exclusively incorporated macroeconomic factors believed to impact the volume of IPOs, while disregarding micro-level company-specific factors. Incorporating micro factors within the model stands to yield more meaningful results.

Implications

This study correlated IPO waves in G-7 countries with specific macroeconomic factors. The analysis revealed that stock returns have a significant positive impact on increasing the number of IPOs. These findings underscore the importance for companies to consider stock market performance in their IPO timing, emphasizing the role it plays in public offerings. Conversely, other macroeconomic variables did not demonstrate a similar effect.

These findings yield a range of significant implications for companies and investors operating in financial markets. Particularly, the necessity for companies to focus on stock returns in their IPO timing is highlighted, enabling a better understanding of factors influencing the performance and competitive advantages of potential IPO companies.

Furthermore, this research may assist in better comprehending the influence of economic conditions and macroeconomic variables on companies’ IPO decisions. This could aid businesses and investors in making strategic decisions based on a more robust foundation in timing their public offerings.

Underlying data

Inflation and GDP growth rate data were retrieved from WorldBank databank and filtered for the G-7 countries for the period of interest (https://data.worldbank.org/indicator/FP.CPI.TOTL.ZG). In the menu on the right of the home page, the indicators for sample countries can be selected. No account creation is needed.

IPO numbers were retrieved from Statista. Scholars don’t have to create account and make subscription on Statista when they are downloading IPO numbers for each country seperately. For multiple countries and periods, the system requires to create an account.

Stock returns were retrieved from Investing.com. In thesearch box located at the top of the home page of investing, equity code is written and profile page of related stock wil be opened. The historical data section is located just above the price chart. After determining start and end dates all information regarding on closing price, opening price, highest and lowest will be visible.

All data used are open access.