Nascent stock exchanges: Explaining success and failure

José Albuquerque de Sousa, Thorsten Beck, Peter van Bergeijk, Mathijs van Dijk 17 November 2016

a

A

Although theory is ambiguous, a large body of empirical research emphasises the importance of well-developed and efficient financial markets for economic growth, at least in developing and emerging economies (Levine 2005, Beck 2013). Many low- and even middle-income countries, however, not only have underdeveloped financial systems, but also have concentrated financial structures, dominated by banks and characterised by the absence of liquid public capital markets. While the search for an optimal mix of banks and capital markets has been so far in vain, there is evidence of an independent effect of banking sector and equity market development on economic growth (Levine and Zervos 1998, Beck and Levine 2004). This, in turn, raises the following question: What explains why some countries have well-developed equity markets while others have shallow and illiquid markets?

In a recent paper, we explore conditions for the successful establishment of public equity markets across a sample of 59 developing countries that have opened a stock exchange since 1975 (Albuquerque de Sousa et al. 2016). Specifically, we use an array of different methodologies to gauge the factors associated with the variation in success and failures of newly established stock markets. We thus complement an expansive literature that has considered cross-country variation in the development of relatively mature stock exchanges. We contribute by shedding light on the early days of new stock exchanges.

How can we measure the success of stock markets?

We can draw on substantial cross-country experience over the past 40 years in setting up new or reviving closed stock exchanges. Since 1975, the number of countries with at least one stock market has more than tripled, from 53 to 165. However, the vast majority of academic studies to date (even the ‘emerging markets’ literature) focuses on at most 50–60 of these 165 countries.

We use three measures of stock market development, widely available and used in the financial development literature:

  1. Market capitalisation to GDP captures the total outstanding stock at the exchanges of a country divided by real economic activity and thus proxies for the size of the stock exchange.
  2. Turnover ratio captures how often the average share changes hands in a given year and is an indicator of the liquidity of the stock market.
  3. Number of firms listed on the stock exchange focuses on the diversification potential of stock exchanges, but also on the importance that the stock exchange has for the real economy.  

Figure 1 illustrates three very different development paths in terms of the number of firms listed across three countries (Czech Republic, established in 1993; Tanzania, 1998; Vietnam, 2003) over the first decade after the establishment of the stock exchange. While Vietnam started with a small number of listed firms, the number of listing grew strongly in the following decade. The Czech Republic started with a big bang, but lost a large number of listing in the following years. The stock exchange in Tanzania, finally, essentially remained dormant in the first decade in terms of listed firms.

Figure 1. Number of listings in first ten years of select nascent markets

Can we clearly differentiate between success and failure?

As a first step to analyse variation in success across markets, we apply cluster analysis to the three success measures for the 34 markets in our sample for which values of the success measures are available in the first 20 years after establishment. Figure 2 plots all 34 nascent markets in this sample along the three dimensions of success as measured after 16-20 years. The x-axis represents the number of listings, the y-axis represents turnover ratio, and the diameter of the circles indicates market capitalisation to GDP. The names of the corresponding countries are depicted in each circle.

We find a clear distinction between two clusters of nascent markets (indicated in different colours). There is a cluster of markets with a relatively high number of listings, large market capitalisation, and high turnover, and a cluster of markets with relatively low values for each of these measures. China and Swaziland are the two extremes along the three dimensions of success. China has the most successful stock market, with an average of 1,477 listed companies (Shanghai and Shenzhen combined), market cap representing 77% of GDP, and turnover of 130% over the period of 16-20 years.[1] Swaziland is the least successful market, with an average of six listed companies, market cap representing 8% of GDP, and almost no trading activity (turnover is close to 0%) in the fourth 5-year interval after establishment.

Figure 2. Successes and failures among nascent stock exchanges

While Figure 2 indicates the success after 16 to 20 years, a similar analysis focusing on the first five years shows that markets with an insufficiently high initial number of listings and turnover fail to make it into the cluster of most successful markets after 16-20 years. On the other hand, markets that start out small in terms of market capitalisation to GDP, but with a relatively high number of listings and turnover from the outset (such as China) can still develop into markets that are successful along all three dimensions of success later on.

 

Is early nascent market success a necessary condition for long-term success?

We use necessary condition analysis, as developed by Dul (2016), to assess the importance of initial success for long-term success. While traditional paradigms of multi-causality presume that each determinant is sufficient to increase the outcome but none is necessary, under the necessary condition analysis paradigm, absence of the necessary determinant results in outcome failure, independently of the value of the other determinants. Our analysis shows:

  • A minimum number of listings and turnover in the first five years are necessary conditions for success along both of these dimensions after 20 years. Stock markets that start out with few listings and low trading activity fail to attract a considerable number of listings and to spur adequate trading activity in a later stage, and run the risk of quickly becoming dormant.
  • There is little evidence that the initial market cap is a necessary condition for long-term success.

What explains success and failure?

To test the effect of different institutional, structural, socioeconomic, and policy factors, we use cross-sectional regressions that relate country characteristics to success 15 years later as well as panel regressions that relate time-variant macroeconomic and other country factors to the success of public equity markets. We find that:

  • The long-term success of nascent stock markets is mostly determined by their early success, and by the environment in which they are established. Stock markets’ early success, as well as country characteristics at the moment of establishment, explain more than 60% of the variation in success 15 years later.
  • The size of the banking sector at the moment of stock market formation is the most reliable predictor of its success. A 1% higher private credit provided by the banking sector and other financial institutions is associated with a 1% higher number of listed companies, a 0.4% higher market capitalisation (% GDP) and a 0.7% higher turnover ratio 15 years later.
  • Panel regressions point to the development of national savings over the life of the stock market as the most important predictor of stock market success.

Conclusions

While our work is silent on the public policy rationale for establishing public equity markets, we shed light on the factors explaining the success and failure of nascent stock exchanges. Scale is an important factor, with a large number of listed firms and sufficient liquidity being a critical condition for long-term success. But it is not only about a sufficiently large number of firms ready to go public and share control with a diverse set of owners, but also investor demand, as seen in a sufficiently large national savings rate that is critical for long-term success of nascent stock exchanges.

In several countries across the developing world, there is a heated debate on whether opening a stock market is sensible when the interest from firms and investors may still be limited. Our analysis suggests that the answer to that question is ‘no.’ Alternatives such as developing the private equity industries, less formal intermediation systems or joining a regional stock exchange can be explored.

References

Albuquerque de Sousa, J, T Beck, P A G van Bergeijk and M A van Dijk (2016) “Nascent markets: Understanding the success and failure of new stock markets,” working paper.

Beck, T (2013) “Finance and growth: Too much of a good thing?,” VoxEU.org, 27 October.

Beck, T and R Levine (2004) “Stock markets, banks, and growth: Panel evidence”, Journal of Banking and Finance, 28: 423-442.

Dul, J (2016) “Necessary condition analysis (NCA): Logic and methodology of “necessary but not sufficient” causality”, Organizational Research Methods 19: 10-52.

Levine, R (2005) “Finance and growth: Theory and evidence”, in P Aghion and S N Durlauf (eds), Handbook of Economic Growth, Elsevier, Amsterdam.

Levine, R and S Zervos (1998) “Stock markets, banks, and economic growth”, American Economic Review, 88: 537-558.

Endnotes

[1] As a comparison, in 2015, the combined number of listings on all US stock exchanges was 4,381, with an aggregate market cap to GDP of 140%, and an aggregate turnover of 165%.

a

A

Topics:  Development Financial markets

Tags:  stock exchange, nascent stock exchanges

PhD candidate, Rotterdam School of Management

Professor of Banking and Finance, Cass Business School; Research Fellow, CEPR

Professor of International Economics and Macroeconomics, Institute of Social Studies, Erasmus University

Professor of Finance, Rotterdam School of Management, Erasmus University

Events