The 28 member countries of the European Union (EU) range from developed to frontier markets. The four largest index provider companies include up to 10 EU member countries as frontier markets, as Table 7.1 shows; the present chapter generically refers to EU 10 frontier markets.

The focus on EU 10 frontier markets was induced mainly by these countries’ position as EU member states and the influence/impact that the European integration process might have over the development of the respective 10 countries’ financial markets. The general expressed idea is that the frontier markets are the next emerging markets and have an important untapped economic potential, providing interesting investment opportunities and diversification benefits (Pop et al., 2013; Chen et al., 2014). Do the EU 10 frontier markets exhibit these characteristics? The answer is a rather complicated one. As of Dec. 2014, the EU 10 frontier markets constituted 3.25% of the EU 28 and only 0.47% of EU 28 equity market capitalization. These figures seem to indicate the growth potential. The EU membership also provides numerous benefits for frontier market economic development, and thus financial development; mainly, the further development of domestic securities markets remains an open question. Eight of the 10 frontier countries come from the former communist bloc, and the reestablishment of their securities exchanges represented an important step toward an open economy. For Malta and Cyprus, the establishment of their respective securities exchanges not until the 1990s was generated mainly by their political situation; both countries gained their independence only during the 1960s. Nevertheless, for all these 10 countries, the national securities exchanges are a source of national pride, as highlighted by O’Hara (2001). As EU member states, all these frontier markets have to face the challenges and the competition pressure generated by the enforcement of the Markets in Financial Instruments Directive (MiFID) in Nov. 2007. The main menaces seemed to derive from the development of alternative trading venues (Skinner, 2007), the potential need to increase the critical size required for an exchange to attract and retain liquidity and to generate the necessary revenues for investments in technology (Haas, 2007), and the possibilities presented to domestic companies to gain access to a variety of trading venues all over Europe. Two reports on Bulgaria and Croatia (World Bank, 2011a,b) show that the alternative trading venues seem not to be of interest for the domestic companies of frontier markets (mainly due to their low liquidity), and the domestic companies’ listing process continues to occur on the main regulated market of European frontier countries. Thus for the moment some of the menaces identified in 2007 seem to surround the EU 10 frontier markets, and their reality and their potential dangers should be considered within any future development strategy of European frontier securities markets.

One of the main important strength of any frontier market is represented by the diversification potential it might offer for international investors’ portfolios. Nevertheless, this advantage is fading away as the globalization of financial markets is accompanied by the integration process of securities (mainly equity) markets around the world and also generates takeovers of domestic companies by multinationals. For the EU 10 frontier markets, the process of integration with the more developed markets of the EU and the takeovers of domestic companies by European multinationals is expected to be more rapid and to impair the advantage of diversification exhibited by these markets for international investors. Under these conditions, a small, less liquid securities market might not be relevant in the EU context, as pointed out by Iorgova and Ong (2008). On the other hand, the continuance and further development of EU 10 frontier markets seem to be needed to support the growth of smaller domestic companies of local and/or regional interest, as stressed by O’Hara (2001) and Andritzky (2007).

The EU 10 frontier markets are confronted with two extremes: the lack of relevance within the EU and the potential evolution toward an emerging market. Therefore an investigation of their development potential is of importance for the academic world and practitioners alike. The academic literature has concentrated until now on the frontier markets’ diversification potential by considering the respective market indices and investigating the integration process with emerging and developed markets. Almost no research has considered the internal diversification potential of the EU frontier markets, a key factor in their future evolution. Moreover, only two papers, by Chen et al. (2014) and Pukthuanthong and Roll (2009), consider all 10 EU frontier markets (included in larger samples) when researching the respective markets’ integration level, while no research so far has been dedicated to only the 10 frontier markets of the EU and their internal diversification potential. The results show that, in general, the EU 10 frontier markets exhibit internal diversification potential and therefore can present interesting investing opportunities for domestic and international investors. The internal diversification potential seems to be lower in the cases of Latvia and Malta, with trading concentrated around the pharmaceutical and banking sectors, respectively. Romania has a peculiar position with an apparently limited internal diversification potential.

Further, the present chapter is structured as follows: Section 2 presents a comparative analysis of the EU 10 frontier markets and the existing research results regarding these countries, Section 3 investigates the internal diversification potential of the EU 10 frontier markets using the Granger causality to establish the existence or absence of a domestic index leader and historical domestic, European, and international betas when only one index is reported for the respective frontier market; Section 4 discusses and concludes.

In the absence of a generally accepted definition of frontier (securities) markets, the following set of characteristics or features was identified based on Miles (2005), Speidell and Krohne (2007), Girard and Sinha (2008), Speidell (2008), Wright (2008), Quisenberry and Griffith (2010), Sadorsky (2011), Speidell (2011), Wawrzaszek and Vadlamudi (2011), Gupta et al. (2012), Lin (2011), and Marshall et al. (2015). This set of characteristics refers mainly to the equity segment of frontier securities markets.

First, frontier markets are small, with low market capitalization and/or low ratios of market capitalization to gross domestic product (GDP). Second, frontier markets are illiquid or have limited liquidity seen mainly in thin, infrequent, and irregular trading activity, and have high transaction costs. Third, the trading activity tends to be concentrated around a small number of securities with significant capitalization. Moreover, the sector weight might favor the financial stocks, which, in some cases, compose the vast majority of the frontier equity market universe. Fourth, frontier (securities) markets represent high-risk investments caused by a wide variety of factors, from those specific to the respective securities markets (eg, the absence or limited availability of short selling, a low level of transparency at market level but mainly at company level, weak corporate governance) to security price volatility, exchange rate volatility, inadequate regulatory frameworks, unsatisfactory protection of shareholders and creditors, a shifting attitude toward foreign investors, political risk, and sovereign risk. Fifth, frontier markets usually present a relatively low level of integration with developed and emerging markets, and also provide greater return potential, as recently highlighted by Porwal (2014) for the S&P’s frontier markets index between 2008 and 2013. Therefore, frontier markets are perceived to be able to provide important diversification benefits for international/global portfolios.

Four of these five characteristics will be discussed based on Tables  7.2–7.4.

As Table 7.2 shows, all EU 10 frontier markets have had about 20 or 25 years of activity. Their small dimensions and relatively low importance in their respective countries’ GDPs is revealed in Table 7.3. Having an average market capitalization of EUR 5759 million and an average ratio of market capitalization to GDP of 22.98%, the EU 10 frontier markets comply with the first characteristic. Two countries, Croatia and Malta, exhibit ratios of market capitalization to GDP similar to those of developed markets—higher than the 25% suggested by Girard and Sinha (2008) as a threshold for developed markets. Thus their market capitalization is low or very low; this phenomenon seems to be characteristic of countries with limited resources and an economic activity dominated by the financial sector. Market capitalization of most of the EU 10 frontier markets follows an upward trend over the 15 years under scrutiny. Romania, Croatia, and Bulgaria exhibit high growth rates, thus continuing to make a modest contribution to their respective countries’ GDPs; an exception is Croatia, with a profile closer to an emerging market. The only frontier market that followed a downward trend is Cyprus, with a slight recovery in 2014.

The turnover ratio was used as a proxy for equity market liquidity. As Table 7.3 shows, the liquidity is low, with an average for all markets for the entire period of 11.26%. Between 2000 and 2002 the liquidity of several frontier markets was relatively high, mainly due to the transactions that followed the privatization programs. After the financial crisis in 2008, most of EU 10 frontier markets have seen their liquidity decreasing, mainly due to the exit of institutional foreign investors. The 10 frontier markets under scrutiny comply with the second characteristic also.

The concentration of turnover around the five most traded companies is also presented in Table 7.3. The average for all countries for the entire period is 72.02%. The countries with a lower concentration around the top five companies are Lithuania, Croatia, and Bulgaria. The country with the highest concentration of trades around the top five companies is Slovakia. Only two countries, Croatia and Malta, exhibit a decreasing trend for this concentration, suggesting that an increased number of companies have started to register important trading volumes. It must be pointed out that the EU 10 frontier markets include two extremes: the Baltic countries and Malta with a relatively low number of listed companies (less than 50 on average for the entire period), and Bulgaria, Croatia, and Slovakia with more than 100 listed companies. While for the countries with a small number of listed companies the turnover concentration around the top five (often representing over 10% of the listed companies) is to be expected, for the second group of countries this concentration indicates that the companies outside the top five are of marginal interest for investors for various motives: from a low free float to lack of transparency and poor levels of corporate governance.

The data for the dominance of the financial sector, as can be seen in Table 7.3, are incomplete. Only for Bulgaria, Cyprus, and Romania were the data extracted from the respective exchanges’ monthly reports. For the remaining seven countries the data were estimated based on the top five concentration; in the case of Malta only two banks (Bank of Valletta and HSBC Malta) were considered. Based on these partial data, only Cyprus, Malta, and Romania seem to have the turnover dominated by financial sector companies, while for the other countries, other sectors like manufacturing and telecommunications seem to be dominant, depending on country specifics.

Therefore, the EU 10 frontier markets only partly comply with the third characteristic.

The risks associated with investments in frontier markets are partly presented in Table 7.2 (sovereign risk and exchange rate volatility risk) and Table 7.3 (main equity indices’ performances). As Table 7.2 shows, 7 of the 10 EU frontier countries, and partly Bulgaria, present medium grades for sovereign risk and being considered as “investable” economies. Croatia is considered speculative and Cyprus highly speculative. The only country that faced an important downgrade is Cyprus due to the banking crisis the country went through during 2011. The ratings for Croatia, Malta, and Slovenia remained almost at the same level as those of 2000, while Bulgaria, Lithuania, Romania, and Slovakia registered an upgrade from speculative or highly speculative ratings.

The exchange rate volatility risk was also relatively low for the majority of EU 10 frontier markets since 2004/05 when seven countries became members of Exchange Rate Mechanism (ERM) II and agreed to peg their currencies to the euro with the further aim of becoming eurozone members and adopting the euro as their currency. Only Bulgaria, Croatia, and Romania present a potential exchange rate volatility risk. However, this should be considered within the frame that these three countries intend at some time in the future to enter ERM II and also to become eurozone members; therefore their national banks currently ensure a certain level of exchange rate stability.

While the the sovereign and exchange rate risks seem somewhat tamed for the EU 10 frontier markets, the price volatility risk, captured by the annual performances of respective securities exchanges’ indices in Table 7.3, tells another story. The average return rate for the entire period for all countries is 12.76%, with Bulgaria and Romania as top performers, while Cyprus is the worst performer of the 10 countries. The values ranging from annual returns as high as 150% and as low as –79% indicate a high level of volatility. Using as a simple proxy for volatility the standard deviation of the annual series, the average standard deviation for the entire period for all 10 countries is 38.03%, which can be considered high. The frontier market with the highest volatility is Cyprus (55.99%), followed by Bulgaria, Romania, and Lithuania with standard deviations higher than 40%, while at the lower end is Malta with a standard deviation of only 25.81%.

The picture of the fourth characteristics is completed by the information in Table 7.4, which shows, according to World Economic Forum ranking, the level of development for the EU 10 frontier markets and the regulation of their respective securities exchanges.

Concerning the financial market development, four countries rank between 25th and 40th position (Estonia, Latvia, Malta, and Slovakia), while the remaining six countries rank below the 50th position, with Slovenia ranking well beyond 100th position. The regulation of the securities markets reveals that only one country, Malta, ranks within or close to the first 25th position, followed by Estonia usually within top 40 positions. The remaining eight countries rank at or well below the 60th position, with Romania having the lowest rank at around 100th position.

The financial development information is completed by the data in Table 7.5, which shows that the EU 10 frontier markets have underdeveloped derivative products and other financial instruments or completely lack those segments, while the turnover concentrates mainly around stock shares and/or bonds, indicating also a low level of market sophistication.

It can be safely inferred that EU 10 frontier markets comply with the fourth characteristic; thus the sovereign risk and exchange rate risk are lower given these countries’ positions as EU 28 member states.

The fifth characteristic is investigated and discussed by a growing body of literature focusing on frontier market integration relative to developed and/or emerging markets. The studies that include a variable number of the EU 10 frontier markets can be split into three categories. The first one incorporates the European frontier markets within medium to large samples of countries (between 15 and 85 countries) and includes the studies of Speidell and Krohne (2007), Speidell (2008), Pukthuanthong and Roll (2009), Berger et al. (2011), Todorov and Bidarkota (2011), Kohlert (2011), Lin (2011), Samarakoon (2011), Guesmi and Nguyen (2011), De Groot et al. (2012), Baumoehl and Lyocsa (2014), Chen et al. (2014), and Marshall et al. (2015). Only the papers of Pukthuanthong and Roll (2009) and Chen et al. (2014) include in their large samples all the EU 10 frontier markets. The second strand of literature is the largest and the oldest, including studies before the countries under scrutiny were labeled “frontier,” and concentrates on smaller samples of countries (fewer than 15) grouped based on regional criteria or on congeniality. Within this category the following studies were considered: Schroeder (2000), Patev and Kanaryan (2002), Mateus (2004), Birg and Lucey (2006), Onay (2006), Syllignakis and Kouretas (2006), Vizek and Dadic (2006), Dvorak and Podpiera (2006), Moore (2007), Samitas et al. (2006), Girard and Sinha (2008), Middleton et al. (2008), Syriopoulos and Roumpis (2009), Wang and Moore (2009), Horobet and Lupu (2009), Allen et al. (2010), Todorov and Bidarkota (2010), Demian (2011), Kenourgios and Samitas (2011), Nikkinen et al. (2011), Syriopoulos (2011), Syllignakis and Kouretas (2011), Gupta et al. (2012), Wang and Shih (2013), Guidi and Ugur (2014), and Reboredo et al. (2015). The third group of studies contains research concentrated on only a number of EU 10 frontier markets and is dominated by the studies dedicated to Baltic countries of Maneschioeld (2006), Moroza (2008), Masood et al. (2010), Nikkinen et al. (2012), and Babalos et al. (2014). One study that concentrated on 8 of the 10 EU frontier markets was found, that of Kiviaho et al. (2014).

Despite their variety as to sample, period, or method of research, all these studies document the partial integration of the EU 10 frontier markets with the developed and/or emerging markets; this partial integration evolved from complete segmentation of the respective countries at the beginning of the 2000s to partial integration influenced by the EU englargement announcement, as documented by Dvorak and Podpiera (2006), but not by the accession, as shown by Demian (2011). Moreover, the integration was shown to vary over time, with periods of increase, decrease, and a new increase during the 2007–09 global financial crisis. Berger et al. (2011) and Guesmi and Nguyen (2011) highlight that the EU member countries and respective southeastern European countries exhibit a higher level of integration than the rest of the frontier markets all over the world. Of the EU 10 frontier markets, Romania and Slovakia seem to be the least integrated, while the Baltic countries have an increased level of integration and therefore offer lower diversification benefits, as pointed out by Nikkinen et al. (2011) and Kiviaho et al. (2014).

Based on the vast number of studies mentioned, it can safely be said that the EU 10 frontier markets comply with the fifth characteristic also; however, this fifth characteristic will have the tendency to become weaker as the integration process is expected to continue.

Hypothesis 1 (when a stock exchange reports two or more equity indices): if, based on Granger causality, no index leader can be found, the stock exchange exhibits an internal diversification potential and therefore might represent an interesting opportunity for investors. This hypothesis is based on the idea formulated by Canegrati (2008) regarding the leading indices at world level.

Hypothesis 1 modified: if, based on Granger causality, the index leader is an all-share/composite index (including a large portfolio of shares), the stock exchange exhibits an internal diversification potential and therefore might represent an interesting opportunity for investors.

Hypothesis 2 (when a stock exchange reports only one index): if for the selected companies the historic domestic beta is higher than the European beta and international beta, the stock exchange exhibits an internal diversification potential and therefore might represent an interesting opportunity for investors. This second hypothesis is based on the idea formulated by Dvorak and Podpiera (2006) regarding historic beta.

Table 7.6 shows the four countries where multiple equity indices are reported and the remaining six countries with just one reported equity index apiece from which three to four companies were selected.