Adoption of global investment performance standards: Case of ASEAN

Abstract Research on voluntary compliance with accepted international standards has paid overwhelming attention to financial reporting standards, but not to investment performance standards. Previous research on the adoption of the Global Investment Performance Standards has overlooked the unique region of the Association of Southeast Asian Nations. Using 17 years (1999 to 2015) worth of data from all ten countries, which generates 170 country-year observations for each variable of the study, this paper evaluates whether, and how, social and economic pressures influence the adoption of GIPS in the region in the Institutional Theory lens. The results suggest that social pressure is more impactful than economic pressure on the adoption of GIPS. The findings have generated useful contributions and implications in this vein, and several future research directions have been identified.


ABOUT THE AUTHORS
The research team has combined professional and academic dimensions. Deborah Zelikson is the first author of the paper, a professional who has decided to conduct her PG studies, including a PG dissertation, motivated by her professional background. The dissertation was jointly supervised by Junjie Wu and Moade Shubita. This research has benefited from Deborah's outstanding dissertation and has been developed jointly by combining the expertise of the three authors.
Moade Fawzi Shubita who is the corresponding author for this paper has published more than 30 articles, mainly in Finance and Accounting journals, in Accounting and Finance topics, and his research interests cover various directions around market-based accounting research areas, primarily using quantitative research methods.
Junjie Wu is a reader of Finance and has published 50 scholarly articles with nationally and internationally excellent journals. Her multidisciplinary research areas span across accounting, finance, economics, energy, international business and education, using diversified research methods.

PUBLIC INTEREST STATEMENT
In a dynamic business environment, uncertainty is the dominant factor, and the need for transparency is no longer an optional dimension as it becomes a necessity by default. Our study contributes to the professional business environment in addressing Global Investment Performance Standards by the Chartered Financial Analyst, as globally accepted standards as proxy for best practice. We have addressed the ASEAN region using country level data, in the hope that our study is used by professionals in ASEAN, as well as elsewhere, in order to motivate the adoption of global standards to help in enhancing sustainability around the globe by showing that there is a social need for this adoption on a voluntary basis with globally accepted investment standards.

Introduction
Investment management professionals manage various assets, such as shares, bonds and real estate, in order to meet specified investment goals for the benefit of the investors. Nowadays, the investment process has become increasingly complex and international, following increases in investment types and the volume of financial entities, and has demanded the standardisation of investment performance calculation methodology, and a presentation of the returns generated. The adoption of a global standard allows investors to quickly compare the investment performance among managers on a "like for like" basis. As a result, firms that are based in countries with limited or no investment performance standards can comply with a globally accepted format in order to enable them to compete for clients on an equal footing with firms from countries with established practices. Besides, firms from countries with developed standards would have more confidence in a fair comparison against local firms in countries with no previous record of standards (CFA Institute, 2014). As such, the adoption of a voluntary standard for investment performance measurement is described as a "global passport", because firms can transport their reports around the world without having to recalculate figures based on local rules (Evans, 2013). The standards also facilitate a dialogue between the client and the firm with regards to the investment performance reports.
The Global Investment Performance Standards (GIPS) are a set of voluntary and ethical principles for investment managers. They have industry-wide application and contain guidelines on how to calculate and present investment performance to clients, adopted by 85% of the top-100 global asset management firms as of December, 2015 (Grover, 2017), representing 61% of the world's assets under management. At country level, 40 countries have adopted GIPS, as of 1 April 2017 (GIPS, n.d.).
Studies on voluntary reporting standards (for example, International Financial Reporting Standards-IFRS) suggest that there is a correlation between adoption and the influence of social pressures (Darus et al., 2013;Lasmin, 2011;Martinez-Ferrero & Garcia-Sanchez, 2017). However, similar studies are rare on the adoption of investment adoption performance standards (for example, GIPS).
Arguably, the concept of a global standard implies one standard fits all countries. On that basis, it assumes that the environment (for example, the composition of organizations, strategies, customs and laws) and the influences upon the environment, for all countries, are the same. Another assumption is that the perceived economic benefits from the adoption of GIPS, from one country to another, would also be the same. However, due to different socioeconomic and political factors, this relationship does not hold for all countries and regions (Lasmin, 2011). For example, the World Development Indicators show that 37% of the countries that have currently adopted GIPS are classified as developing, and 63% as developed countries (The World Bank, 2016). A counter-example is that the European Union (EU) and the ASEAN are two significantly different regional economic organizations (Martell, 2010); however, they have the same 40% rate of GIPS country adoption (GIPS, n.d.). This observation suggests that the economic drivers may not solely determine the adoption of GIPS for a given country, and alternate factors could also influence GIPS adoption, such as the level of sophistication of education programmes, technical challenges, local regulatory bodies for endorsement of standards, and such like. Also, in terms of the profile of an investment management firm or country of origin, its environment can be directly influenced by sociological factors, including religion, ethnicity, economic status, education, location, and political systems. As pointed out by Friedland and Alford (1991, p. 232), "It is not possible to understand individual or organizational behaviour without locating it in a social context".
Since its foundation in 1967, ASEAN has been in the process of regional integration, but the integration has not yet led to tangible results because of the fact that the region fails to address the necessity for legal change (Orcalli, 2017) due to significant diversities across member countries in local preferences, cultural sensitivities, language, and religion (Vinayak et al., 2014). From an investment management perspective, the diversities suggest that a one-size-fits-all strategy might not be appropriate across the varied markets. However, an alternative argument is that if some ASEAN members have adopted an international standard, such as GIPS, then the adopting countries in the region will influence and encourage the non-adopters in order to consider GIPS compliance. This argument links to the Institutional Theory, which suggests that over time, firms within a field become isomorphic, that is, similar in terms of structure, processes, and the way they react to social influences (DiMaggio & Powell, 1983;Nyahas et al., 2017;Tolbert & Zucker, 1983). Similar studies have leveraged DiMaggio and Powell (1983) three categories of isomorphic change mechanism (coercive, mimetic and normative forces) in order to represent social pressures. Coercive pressures relate to formal and informal influence by legal and regulatory bodies; mimetic effects occur in times of uncertainty, and normative pressure stems from the professionalization of a field.
The influence of social pressures on the adoption of investment performance standards is a topic that remains overlooked.
This study, therefore, examines the impacts of social and economic pressures on the adoption of GIPs in the ASEAN region. The analysis is to achieve the following: (1) determining whether social pressures, represented by coercive, mimetic and normative isomorphic change predictors impact the adoption of GIPS; (2) examining whether economic pressures impact the adoption of GIPS; (3) ascertaining whether social pressures are greater than economic pressures on the influence of a country's adoption of the GIPS; and (4) establishing within social pressures which component is more influential than others in the adoption of GIPS. By achieving these objectives, this study would be the first research of its kind to be conducted in the region on this topic. We are determined to shed light on the issues related, such as the current status of the adoption of GIPS, the roles that social and economic pressures have played in the adoption of GIPS and why, and what the region's policymakers should do to improve the adoption rate, and what future research can be carried out in order to deepen the understanding of this topic.
The rest of the paper is structured as follows: Section 2 introduces the adoption of GIPS in the context of the ASEAN. Section 3 provides a theoretical framework for sustainability reporting. Section 4 establishes the linkage between the empirical literature review and the hypotheses. Section 5 introduces the data and analysis model. Section 6 presents the results, and the final section (Section 7) summarises and concludes the paper.

Global investment performance standards within the ASEAN context
The Association of Southeast Asian Nations (ASEAN) was established in 1967 by the original five members (ASEAN-5), Indonesia, Philippines, Singapore, Thailand and Malaysia, and has now extended to ten member countries with a population of more than 650 million. It aims to integrate the region as a whole with the motto "One Vision, One Identity, One Community" in order to achieve the sustainable development of this region, and to compete in the world. After decades of effort, ASEAN has become an investment icon within global financial markets, and is the fourth largest trading region (Thampanya et al., 2020).
Regarding GIPS adoption, the ASEAN has a mix of status: there are four fully adopted GIPS countries of the ASEAN-5, excluding Malaysia. Registered interest countries include Malaysia and Brunei Darussalam, as well as the non-adopters (Cambodia, Lao PDR, Myanmar, and Vietnam) (GIPS, n.d.). Compared to other regions, the ASEAN is behind in adoption GIPS. The Americas continue to dominate the adoption of GIPS globally, with more than 81% of firms claiming compliance from the region. Europe, the Middle East and Africa (EMEA) account for 14%, and the remainder are from the Asia Pacific (Grover, 2016). Grover (2016) further reported that for the 2015 notification period, only 78 of 1,621 firms claiming GIPS compliance were based in the Asia Pacific region (not down to the ASEAN as no breakdown figures are available). This number increased for 2016 with 94 out of 1,608 1 firms claiming compliance in the region (Grover, 2017). The CFA Institute indicates that these figures reveal an opportunity for stronger outreach efforts in the EMEA and Asia-Pacific (including ASEAN) where the GIPS standards are currently underutilised.
There are three possible reasons as to why the ASEAN is under-represented in GIPS adoption, as discussed in the literature. The first relates to the investment firm scale in the region. Many of the local asset managers are classified as small scale. Investment firm scale is measured by assets under management (AUM), which is recognised as a limiting factor for GIPS adoption. The relationship between GIPS adoption and investment firm scale is evidenced in Healey's 2014 report, with the survey indicating that 40% of boutique investment managers agreed that the compliance with regulation becomes the biggest obstacle to expand their client base. The second reason may be due to a limited number of global firms locating their headquarters in the region (usually only having their "Asia branch" offices located in Singapore or Malaysia). The third reason is attributed to the fact that some developing countries of the ASEAN (that is, Cambodia, Laos, and Myanmar) have minimal exposure to international standards due to their immature legislation environment (Grover, 2016(Grover, , 2017. Grover (2016) also argued that 2015's figures provided a starting point for conversations between GIPS and the local regulators, and the further promotion of standards in the region. With the scope for new adoption in the ASEAN countries, it would be valuable for the research to help understand what might influence this adoption process.
Moreover, it is notable that the ASEAN was relatively unscathed from the 2007-09 Global Financial Crisis due to the measurements taken after the Asian Financial Crisis of 1997-98 (Chia, 2011). This phenomenon subsequently encouraged economic policymakers and promoted investment benefits in the region. In March 2012, the ASEAN Investment Area (AIA) Council introduced the ASEAN Comprehensive Investment Agreement (ACIA), setting out the strategic objective as a "transparent and competitive investment environment" (ASEAN, n.d.a), which has resulted in many innovative projects with the unique feature of all being underpinned by the theme of inclusion in the investment management field. This social context would encourage and foster the adoption of global voluntary standards for investment management and make the ASEAN a fitting case/region for research on the adoption of GIPS.
3. Theoretical framework for sustainability reporting 3.1. Institutional theory and its representativeness of social pressure Institutions are defined as "regulative, normative and cognitive structures and activities that provide stability and meaning for social behaviour" (Scott, 1995, p. 33). Institutional theory concentrates on the crucial question of why organisations in a field, over time, generally look/act the same (Miles, 2012). At the beginning of the organisational lifecycle, there is significant variety and form; however, there would be homogeneity in the organisational configuration and practices in the mature stages (DiMaggio & Powell, 1983). Early institutional scholars recognised the notion as institutional isomorphism, whereby processes and structures of organisations evolve and become similar to one another (DiMaggio & Powell, 1983;Meyer & Rowan, 1977). They suggested that organizations are homogeneous in composition because they need to gain a social mandate or legitimacy, and they do this by adhering to social expectations. Organisations not only begin to look alike but also respond similarly to regulatory, normative pressures and situations of uncertainty (DiMaggio & Powell, 1983).
According to DiMaggio and Powell (1983), the diffusion of ideas, standards and recommended structures facilitate the process of institutional isomorphism. Their framework covers the mechanisms through which institutional isomorphism change occurs (Boxenbaum & Jonsson, 2008). This approach involves testing the isomorphic pressure(s) to account for the diffusion of a standard or practice. The theory posits that three change mechanisms make institutional isomorphism happen by exerting pressure on organisations: coercive, mimetic and normative.

Coercive isomorphism
Coercive isomorphic change is the result of regulative organisational elements that emphasize compliance as the basis for legitimacy. It applies when new policies, government mandates and financial reporting requirements are implemented. DiMaggio and Powell (1983) defined coercive isomorphism as resulting from formal and informal pressures exerted on organisations. Pressures from stakeholders and expectations from society persuade organisations to conform (Palthe, 2014;Peng, 2002). Despite empirical studies have been divided on the variables used to represent coercive isomorphic pressure (Lasmin, 2011), researchers agree in general that the adoption of global policies and standards is a kind of coercive isomorphic change (Henisz et al., 2005;Weber et al., 2009).

Mimetic isomorphism
The mimetic isomorphic change focuses on cultural legitimacy. Organisations face uncertainty in times of making a significant change which could result from inadequately understood technologies, ambiguous goals, and unclear relationships (Henisz et al., 2005). In this case, shared models and beliefs are invoked and preferable (Palthe, 2014). An organisation may seek credibility by imitating the entities perceived as the most successful and prominent in their field (DiMaggio & Powell, 1983). By replicating those organisations that have already attained cognitive legitimacy, new entrants can also be deemed legitimate (Suchman, 1995).
Mimetic pressure was divided into three classifications: frequency-based, trait-based or outcome-based. Frequency-based imitation refers to the influence generated when a large number of organisations has adopted a structure or practice in the population. Trait-based imitation is when organizations look to replicate highly successful peers. Outcome-based imitation occurs following evaluation of the strategy of others. Outcome-based is limited to highly visible events, such as signals from the capital markets (Henisz & Delios, 2001).

Normative isomorphism
Normative isomorphic change is associated with the immediate environment of the organization and moral legitimacy. Compliance with familiar structures and policies of a field stems from professionalism (DiMaggio & Powell, 1983). Examples of change mechanisms include networks that span the organisational area, inter-hiring between firms, andalso educational accreditation. Over time, professionals within a field, with a similar knowledge base and network, address issues in the same way (Palthe, 2014). Suchman (1995) defined the pursuit of professionalism as organisations "linking their activities to external definitions of authority and competence". The moral values imposed by a culture upon organisations also influence the decisions regarding what is an acceptable strategy (Martinez-Ferrero & Garcia-Sanchez, 2017, p. 589). Therefore, adherence to widely respected ethical standards and codes of best practice can have a similar effect and impact how an organisation operates (DiMaggio & Powell, 1983).
The claim to a supply of knowledge and technical specialists is a powerful source of credibility and a driver of normative isomorphic change identified by scholars (Hopwood, 2000;Power, 2003). Hopwood (2000) suggested that the adoption of standardised financial accounting practices or similar is not only functional but also symbolic of modernity and expertise. Furthermore, evidence supported a correlation between the enhancement of an organisations status and the membership of professional bodies and expert panels of the senior management (Abbott, 1981;Suchman, 1995).
In summary, pressures to make substantial changes on a specific process that results in a behavioural outcome in an organisation, a sector, a country, or a region, through political and legislative influence (coercive), replicating success (mimetic) and standardized professional norms and benchmarks (normative). All three isomorphic change mechanisms are therefore characteristic of a form of social pressure.
Until this point of the section, we have demonstrated social pressure can be represented by institutional isomorphism. In the next section, we move to develop hypotheses to test data based on empirical evidence on the relationships between social and economic pressures, respectively, and the adoption of GIPS.

Social pressure and the adoption of GIPS
Researchers debate the concept of institutional context and refer to it as "rules, norms and ideologies" (Meyer & Rowan, 1977, p. 84), a regulatory framework of government and professional bodies (Meyer & Scott, 1983), or social setting and symbolic/cultural influence (Tolbert, 1985). However, it generally relates to broader social understandings and associated pressures, as well as the adaptive behaviour by organisations to secure an appropriate "fit" to the environment facing and the rationale behind it.
In the context of voluntary standard adoption, empirical studies are observed more on, for example, financial reporting standard but scarce on investment management standard such as GIPS. Arguably, they should share common ground in terms of globally accepted professional standards. These studies established a positive relationship between social pressure and the adoption of voluntary standards in financial reporting, e.g., Lasmin (2011) presented a correlation between all three isomorphic change mechanisms and the adoption of the IFRS. Similarly, a positive relationship between coercive, mimetic and normative pressures and assurance of sustainability reports is found in Martinez-Ferrero and Garcia-Sanchez (2017). They argued that voluntary adoption acts as a legitimisation tool; that is, countries that have a more significant legal system, and cultural development, are more likely to adopt voluntary standards. In contrast, research also acknowledged that despite coercive, mimetic and normative pressures drive compliance with IFRS, "a logic of resistance" was identified to the introduction of new standards (Maroun & Van Zijl, 2016) because new requirements are not perceived as competent. Therefore non-compliance by some stakeholders having negative sentiment is inevitable.
Although the GIPS Standards have been in circulation since 1999, less empirical evidence exists on GIPS studies. It could argue that the standards are now in the latter stage of the diffusion lifecycle since part or all of 23 of the 25 largest asset management firms globally claim compliance (Kent, 2016). In other words, this suggests that social pressure on the adoption of GIPS matured. For instance, the CFA Institute, as with any standard-setting group, is required to be aware of the social pressures influencing their body of work. At a country level, social pressures comprise coercive, mimetic and normative forces. Coercive force involves local organisations to develop and promote GIPS in partnership with the CFA Institute (Langevoort, 2014) 2 ; mimetic force implies that the size of capital markets that represents the openness of the country's economy (Touron, 2005), which in turn may influence a country's willingness to comply with GIPS or replicate the processes of a perceived prosperous country; and normative forces are presented based upon the levels of higher education and the professional networks in the region. It can argue that the greater the influence of social pressures, the higher the adoption rate of GIPS. From this reasoning, the first hypothesis proposes: Hypothesis 1: Social pressures, represented by coercive, mimetic and normative isomorphic change predictors, will exhibit a positive relationship with the adoption of GIPS.

Economic pressures and the adoption of GIPS
Economic pressures define as changes in the business environment, such as the globalisation of markets and competition (Pirinen, 2005). A positive growth rate in GDP is a signal of expansion and good economic health. Increases in foreign direct investment, as well as domestic companies' financing activities abroad, also serve as an indicator of a countries credibility and financial position (Lasmin, 2011). At the firm level, there is pressure to maximise the wealth of its shareholders, measured by the rate of the earnings per share, with the financial decisions being influenced by factors that would increase the net present worth of a company's profits (Rappaport, 2006). For many standard-setting bodies, economic pressures often see to be the driving force of dissemination of the standard (Martell, 2010). Pirinen (2005) examined the adoption of International Accounting Standards (IAS) in Finland and found that economic pressures were the leading cause of the change in financial reporting in line with the international practice since the end of the 1970s, supplemented by social pressures. Another study carried out by Chi et al. (2013) in Taiwan revealed a compelling case: before 2001, under the Taiwan Companies Act, firms with contributed capital above a certain threshold had to produce financial reporting. This mandatory requirement was rescinded in 2001, leaving the voluntary reporting as an incentive to meet listing rules at the two local stock exchanges. The result shows that a continuation of financial reporting post-2001 correlated with the desire to obtain external financing-an excellent example of economic pressure impacting on the adoption of voluntary standards. Additionally, studies also found, while economic incentives are necessary, the quality and usefulness of reporting standards adoption would be high (Ball et al., 2003).
In the professional sector, Greenwood et al. (2002) studied accounting firms and concluded that by framing the profession, as being "under threat", it was portrayed that the way to stay competitive and meet new demands was to accept change by compliance with new standards. Taking investment management profession as an example, while it is not currently in the midst of a financial crisis as seen in 2007-09, the lack of transparency and risks control continue to call out as a barrier to trust (Edelman, 2016). This, in turn, has an economic repercussion for the sector, because asset owners would think twice before placing their investments. As such, it might be a good strategy for investment management firms to adopt best practices and seek compliance with standards such as GIPS to acquire new and retain existing business. This perspective can lead to the second hypothesis: Hypothesis 2: Economic pressures demonstrate a positive relationship with the adoption of GIPS.

Social vs. economic pressures and the adoption of GIPS
Our hypotheses 1 and 2 discuss the relationship between social, economic pressure respectively and the adoption of GIPS. This relationship is universal. We now would like to explore further to see social or economic pressure, which has more influence than the other in the adoption of GIPS. It could argue that the relationship should discuss in the context of a country or region in a certain period. Our case is ASEAN.
We have noticed that most of the previous studies on ASEAN regions have addressed issues on a country level, for example, Widyaningsih et al. (2019) study about the impact of audit rotation on audit quality in Indonesia. Adi et al. (2016) analysed the quality of performance report for local governments in the same country, while Nguyen's (2015) study addresses the lending rate in Vietnam following the financial crisis. Nor (2016) discusses the nature of tax planning in Malaysian multinational companies.
Despite both social pressure and economic pressure having a positive relationship with the voluntary adoption of global professional standards, research revealed in a certain period, and the two forces could conflict, e.g., Lasmin (2011) concluded that country adoption of IFRS is pressured more by social legitimacy in the form of isomorphic change than by economic pressures. The tension between economic and social influences can even provide insight into how global standards have been developed (Rodrigues & Craig, 2007).
In the context of ASEAN, the ASEAN Economic Community (AEC) was established in 2015, aiming to achieve an integrated and cohesive regional economy. The leaders of the member states adopting the AEC Blueprint 2025 (ASEAN, n.d.b). The framework sets out steps to establishing a competitive single market and the free movement of goods, services, capital and investment between member countries. Studies agree that the ASEAN economy has experienced significant growth since 2000, embracing globalisation and attracting foreign investment (Vinayak et al., 2014;Yean & Das, 2015). However, critics noted that the AEC integration timetable has been a struggle for the region, with many objectives unfulfilled from the original 2007 plan and carried forward in the Blueprint 2025 (Lehmacher, 2016;Yean & Das, 2015). Some authors argued that the region's integration requires more open to the outside countries (Pangestu & Ing, 2016), such as establishing a partnership with non-ASEAN countries (Chen et al., 2017). One reason suggested for the slow progress is that regional responses to economic influence could only proceed with the consensus of all ten national governments (Lehmacher, 2016). A shared vision is difficult to reach because each ASEAN country is very different and have dissimilar priorities. For example, Indonesia is a member of the G20 representing nearly 40% of the region's economic output. In contrast, Myanmar is considered a frontier market that is emerging from decades of military control and civil war (Vinayak et al., 2014). Similarly, it could argue that external economic influences would have limited impact, due to the lack of consensus on the appropriate reaction from the region as a whole. Yean and Das (2015) suggested that resolution of conflicting domestic issues was a prerequisite to deeper integration and subsequently a cohesive regional response to economic pressures.
In contrast, the region's response to social pressures displays a strong cohesion, in the spirit of the ASEAN Way that has origins from tradition and culture because a key objective of the Socio-Cultural Community is to preserve a regional identity. The ASEAN Regional Forum plays a role in facilitating international relations and coordinating a single holistic response from the ten members (Laurence, 2007). Evidence shows that those countries with more power in the global system, or possessing high status have shaped the policies for countries that are less powerful or considered less legitimate (Gilpin, 1987). This explanation could apply to Singapore and Indonesia, in so far as other GIPS adopting countries influenced them in the investment management global community such as the UK and Japan. Likewise, within the ASEAN region, it could also be proposed, that the GIPS adopting countries will, in turn, exert the primary influence on the nonadopters. This inference leads to the argument that social pressures more impact the ASEAN region because the consensus among the member states is more prevalent than in response to economic pressures. From these discussions, the third hypothesis developed: Hypothesis 3: Social pressures have more considerable influence than economic pressures in a country's adoption of GIPS within the ASEAN region.

Social pressure components and the adoption of GIPS
In the studies of social pressure representing by coercive, mimetic and normative isomorphism, all three processes are embedded in different origins of how the behaviour of institutions diffuses. It suggests researchers do not give equal weighting to all three components, whereas being selective in interpretation, e.g., Tolbert and Zucker (1983) tested the influence of coercive pressure against mimetic pressure and found coercive pressure to be more effective in the spreading of new administration reform. In contrast, Lasmin (2011) confirmed that all three of the change mechanisms jointly influence accounting standards adoption internationally, but insufficient evidence supports an order of significance. The adoption of accounting standards by US firms highlighted normative pressure as the most important change mechanism (Mezias, 1990). In the study of Martinez-Ferrero and Garcia-Sanchez (2017), normative pressure was also the factor that provided the most significant explanatory power in the voluntary assurance of sustainability reports and followed by coercive pressure as a secondary stimulus. These results indicate that the evaluation of each social pressure component's influence on voluntary standard adoption, such as GIPS also needs to link to the institutional environment in the adoptive country or region. Some studies identified that coercive pressures are likely to lead to ceremonial adoption of a structure or process because the adoption is not backed by locally embedded skill or motivation (Weber et al., 2009). Substantive adoption is likely to occur when internal knowledge and experience support the development and refinement of a practice (Meyer & Rowan, 1977) while professional networks can act as transmission channels for organizational models and technical knowledge (Owen-Smith & Powell, 2008). This finding is consistent with a primary objective of the GIPS Executive Committee, i.e. to partner with local sponsoring organisations, to promote expertise further and educate professionals on the standards. This driver of professionalism is characteristic of normative pressure. Concerning the context of ASEAN, we argue that normative pressure plays more roles than coercive and mimetic pressure. The evidence followed regarding the prospect of regional education, and the foster of professionalism and networks, and the increase of investment management research in the region can support our argument.
The initiation of ASEAN Vision 2020 calls for investments in the development of a knowledge economy. A case study of higher education in Malaysia and Thailand found that university interviewees placed significant importance on the international university rankings (Chapman & Chien, 2014). High rankings view as a way to attract top talent, foster professionalism and earn global respect. Approximately one-fifth of international students in Malaysia and Thailand are from ASEAN member states. Based on the assumption that graduates will stay in the region, it follows that the influence of educated professionals and networks will exert normative pressures on organisations to keep current with standards.
The CFA Institute reported that in the wider Asia-Pacific region, 11 of the 19 local member societies requested brand campaign funding (CFA Institute, 2016). In the ASEAN region, brand campaign initiatives planned for cultures in Indonesia, Philippines, Singapore and Malaysia. The campaigns based on out-of-home, digital and print advertisements, e.g., posters in the local commuter train stations and Facebook competitions. The Global Brand Campaign aims to deepen the understanding of the professionalism agenda and position the CFA Institute as the professional body for the investment management industry (CFA Institute, 2016). Research gathered from Scopus (a multidisciplinary database of journals and conference papers) revealed a growth in publications in economics, econometrics and finance for the region (Halevi et al., 2014), with a growth rate of 14% in terms of publications for the period 1997 to 2012, the discipline ranked joint 5th out of 26 subjects. This evidence shows a strong emphasis on professionalism, education and networks of talent in investment management-related subjects, as an essential influence that would support the adoption of GIPS. Besides this, there is little indication of a leading coercive pressure from regulators or capital suppliers concerning GIPS adoption. Also, there is no example of a standout investment manager in the region that is championing GIPS and would act as a motive for mimesis. Therefore, normative pressure appears to rank the highest, of the three isomorphic change mechanisms in the region with regards to GIPS adoption. Our final hypothesis arrives with this line of reasoning: Hypothesis 4: Normative pressures are more influential than coercive and mimetic pressures in the adoption of GIPS in the ASEAN region.

Research design
The primary objective of the paper is to examine how social and economic pressures impact the adoption of GIPs in the ASEAN region. The data used covers all 10 ASEAN countries spanning 17 years (1999)(2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015) for 170 country-year observations for each variable. The data collected from the World Bank database and the Chartered Financial Analyst (CFA) Institute. An effort was made to ensure the accuracy of the data before the analysis, various sensitivity analysis and control variables were added to secure the stability of results.
The 17 year period of the study allows the researcher the flexibility to run the study while minimising bias in comparison, due to the different timings of adoption, or expression of interest for different countries.

Data description
The dependent variable (the adoption status of the GIPS) collected from the CFA institute online records. We measure it as a binary variable, where zero is "no adoption" for the country-year observation, and one indicates "a full adoption" or "expression of interest". Table 1 demonstrates three levels of adoption (full adoption, non-adoption and expressed interest of adoption). We use only two levels in the analysis because the expression of interest is a transition period before full adoption for almost the case of the full adopters. The information in Table 1 was gathered on 1 January 2017 but data from previous years show a different pattern, as revealed in the frequency analysis of the adoption status. (2011), we do not want to limit our study to a single period as we want to reflect the behaviour over the entire period of the study. Moreover, to the best of our knowledge, this is the first study to be conducted in this region. Our aim is not limited to testing hypotheses, but to present to readers more information for the whole period in this region. Therefore, we generated a frequency analysis for the adoption status using three levels from 1999 to 2015 for the entire country-year observations. The results reveal that the frequency (percentage) of full adoption status as 14 (8.2%), expressed interest as 38 (22.9%), and no adoption as 117 (68.8%). When rearranging the data per year, we can see that full adoption began in 2007, and for all four full adopter countries. The process was preceded by the expression of interest for different periods, beginning in 1999.

Unlike Lasmin
We argue that expressing interest led eventually to full adoption; this argument is valid during the period of the study for four out of five countries (except Malaysia). Meanwhile, in Brunei Darussalam, interest expression was made in 2014; thus leaving only a short period in which to convert to full adoption during the study. 3 Therefore, to facilitate better representation and in light of the results, we retain the adoption status as a binary variable on two levels. "Zero" is non-adopters having 117 observations representing 68.8% and "One" is adopter (fully or partially; expressing interest) having 53 observations representing 31.2%. This is the dependent variable (binary variable). The description of the other variables is to address in the following sub-section. Table 2 presents the proxies and measurements of each variable and the sources of data. As indicated previously, all proxies have utilised in the literature. Further discussion about each proxy can see during the analysis of the results; a few more control variables were added to the model to capture any other impact from other resources and minimise the misspecification. Those variables will discuss in greater detail in the robustness analysis.

Analytical model and supporting statistics
Alongside testing the hypotheses, we intend to provide readers with additional analyses to reveal more indicators to help future studies and enhance our contribution to the literature. The primary model runs as Logistic Regression testing for the hypotheses to address research objectives. Meanwhile, other analysis methods (e.g., description, correlation and Cross Tabs) adopted to provide supportive statistics to the main Binary Logistic model. After running the model, robustness tests are performed to ensure the trend of results will remain the same.
Given the fact that the response (Dependent) variable is a binary variable (0,1), Logistic Regression model is justified (Field, 2013, p. 761). 4 It is worth noting that as logistic require the dependent variable to be categorical which in our case is binary, it places no restrictions on the nature of the independent variables. Different measurement scales can be combined within the model among the independent variables (Field, 2013).
Where:  6. Empirical results and discussion 6.1. Descriptive data Table 3 presents mean, median, and standard deviation for the main variables. With regard to adoption status, the dependent variable in the model, the mean is 0.31, which confirms that there are 53 adoption observations out of the 170 valid observations, representing 31%. The median is 0 since more than half of the study population did not adopt GIPS during the study period; we anticipate this percentage will decline in the future based on the data trends. While this result needs to be treated with caution since the variable is binary and there is a time lag in the adoption of GIPS, the majority remain in the non-adopter category. However, we computed the percentage of the adoption including the year 2016 and found that the adoption status increased to 33%. Moving forward, we expect the rate to grow based on the trend of the figures of this variable. Thus, for future studies, it will be worth observing how our results would compare.
As a reflection of the variations among ASEAN countries, there are variations in most of the study variables. For example, Market Capitalisation varies among ASEAN countries, generating a high variance. Moreover, the number of CFA varies significantly among countries ending in very high variations. This suggests adopting a technical method of creating quantiles to reduce the variability while retaining the nature of the variable. This approach has been taken in several studies. For example, in her seminal paper, Dechow (1994) converted different variables into quantiles and ran the model per quantile. This issue has been addressed later when in the model analysis.

Correlation and logistic regression
As a further step of exploring the data, we have generated a Pearson and Spearman correlation table with the adoption status. Table 4 presents the correlation coefficients for both Person and Spearman coefficients with the adoption status.  From the results presented in Table 4, while both figures generated by Pearson and Spearman are having the same trend, there is a noticeable difference for the number of character CFA holders, reflecting the high variation, as detailed in Table 3. Closer figures are revealed for the two tools when we have created a quantile variable from the raw figures, as depicted in the row below the same variable in Table 4. We intend to use the quantile version of CFA holders as the normative proxy since it demonstrates more excellent stability than raw figures. The model will be run based on the original version for comparability in the robustness analysis section.
Correlation figures reveal a significant relationship with the adoption status; it should be noted that the sign was unexpected for two variables (ODA and GDP Growth). However, caution needs to be exercised here as correlation figures should not be used to suggest causality or as an indicator of the relationship direction since other factors may have led to such figures (Field, 2013, p. 270).
Logistic Regression is used to test the hypotheses and explore the main factors that may explain the adoption of GIPS. Three stages are undertaken: (1) testing multicollinearity problem; (2) testing the significance of the model, and (3) testing the coefficients of the independent variables in the model as a tool for testing the corresponding hypotheses. Table 5 presents the model using Logistic Regression. It is divided into four panels (see explanation shortly). However, before testing the hypotheses based on the model outcomes, a multicollinearity problem must be examined. The figures required to detect such an issue are presented in Table 5. 6 If tolerance value is less than 0.10, which means the Variance Inflation Factor (VIF) is more than 10, there will be a cause of concern for multicollinearity problem (Myers, 1990as cited in Field, 2013. In Table 5, all tolerance values are higher than 0.1 and even higher than 0.2. This means VIF amounts are less than 10, and even less than five. This shows there is no serious multicollinearity problem that may affect the model of the study outcomes. In other words, the interactions among independent variables are not strong enough to create noise that leads to model misspecification. After examining the multicollinearity and ensuring there is no serious problem in the model, we can proceed to examine the overall model before exploring each hypothesis in the same order of the hypotheses development section. 95% confidence level will be applied (a typical in Social Sciences studies). However, any results remain significant at 90% for weaker outcomes, or 99% for more reliable results will be commented on. Table 6 presents the final version of the Logistic Model, separated into various panels for simplicity of discussion.  The cut value is.500. Zelikson et al., Cogent Business & Management (2020), 7: 1746169 https://doi.org/10.1080/23311975.2020 In Table 6, Panel A reveals that the overall test model is significant at 95% and above; to test the model, we look at Omnibus Test of Model Coefficients where the output generates Chi-Square value that equal 132.10 with Sig. (P-Value) 0.000. This indicates that the model is statistically significant as P-value is 0.000, which is less than 5%. From the results presented so far in Tables 5  and 6, Panel A. It illustrates that there is no significant multicollinearity problem and the model as a whole is highly significant. In Table 6, Panel B reveals that the Nagelkerke R-square, an equivalence of R-Square in the Logistic Regression, is equal to 0.76 (76%). While Panel C in the same table shows that the overall percentage (hit ratio) of the model correctly classifying the voluntary adoption of the GIPS is 86.5%. More importantly, the significance of the model and the absence of significant multicollinearity enable authors to use the model to test the study hypotheses as follows. Panel D of Table 6 presents the results required for testing the study hypotheses.
Regarding the first hypothesis, there are three proxies for coercive, mimetic and normative. In Table 6, Panel D shows that the P values for Net Official Development Assistance (ODA), Market Capitalisation (MC), CFA are 0.344, 0.075, and 0.001, respectively. This indicates, at our confidence level, only CFA as a proxy for the normative dimension is significant. The significant levels of coefficients are for MC and the number of CFA holders, while the net ODA was insignificant. Regarding the direction of the relationship, the two significant coefficients are both positive since the EXP (B) for MC and CFA are 1.014 and 5.793 respectively. Further, we have generated the 95% confidence interval (CI) for Exp (B) for all variables, and it lower and upper bounds are 1 and above for CFA; thereby indicating a positive relationship significant at 95% CL. Meanwhile, the lower bound of MC is close to one (0.999), but it is significant at 90% CL. In figures, the 95% CI for MC = [0.999, 1.029] while for the CFA [2.070,16.215] and for net ODA [0.907, 1.035]. The lower bound of net ODA CI was less than one and equal 0.907, and this coincides with the Exp (B) for net ODA = 0.969 showing negative direction, but this cannot rely on since it is insignificant as explained earlier and the upper bound is more than one equal 1.035.
To conclude, the first hypothesis is valid for two indicators which are MC and CFA at 90% confidence level, and only one indicator which is CFA at 95% and even at 99% given its P-value is 0.001 while there is no significant evidence for the ODA due to the high P-value. This leads to the conclusion that there is a moderate impact for the MC variable as the proxy for mimetic while the normative proxy has generated the highest confidence level. However, we leave the comparison for the third hypothesis.
For the second hypothesis, the results presented in Table 6 reveal that the parameter for the GDP growth as a proxy for the economic dimension is insignificant, P-value = 0.184, which indicates that this is an insignificant result at 90% confidence level. Further, the Exp (B) for GDP less than one along with the lower bound of the CI that equal 0.675. While the figures suggest a negative direction for the relationship, this cannot be relied on given the fact the outcome for this variable is insignificant. In conclusion, there is no evidence to suggest that GDP as a proxy for the economic dimension could have a positive impact on the attitude of adoption or expressing interest of GIPS. Thus, this argument provides no empirical support for our study.
The remaining two hypotheses are an extension of testing the first two hypotheses. A further comparison between social and economic pressures in Hypothesis 3 will be carried out, and among social pressure components in Hypothesis 4.
Hypothesis 3 suggests that social pressures are expected to have greater association with the adoption of GIPS than economic pressure measured by GDP growth. The fact that two of three social power proxies were positively significant at 90% as discussed in Hypothesis 1, and the GDP is insignificant as revealed in Hypothesis 2 make this issue clear and lead to the conclusion that overall social pressures are stronger than economic pressure in influencing adoption of GIPS at the country level in the ASEAN. As GIPS relates primarily to financial statements of the companies, macroeconomic factors are less important compared with Capital Markets and professional dimensions as a proxy for mimetic and normative factors, respectively. To further support this outcome, the amount of Exp (B) for GDP was the lowest compared with other variables; Panel D of Table 6.
Moving forward in the comparison analysis as indicated in Hypothesis 4, we argue that the normative factor is higher than the other two social pressure factors in influencing the adoption of GIPS. The results presented in Table 6, Panel D reveal a definite advantage to the normative pressure measured by the number of the CFA holders, the EXP (B) for CFA equal to 5.793, which is higher than the EXP (B) of MC that equal 1.014. Further, the significance level of the normative factor proxy is lower than the significance level of MC. Thus, the confidence level of CFA is higher. In this hypothesis, the CFA is significant at 90, 95, 99% based on its P-value that = 0.001 while the MC is substantial at 90% only since its P-value 0.075. This yields sufficient evidence to accept this hypothesis, which complies with the literature. Moreover, we can conclude that the normative pressure is a higher factor of social pressure. This also reveals the value relevance for having more professionals in the ASEAN countries and encouraging other countries with low levels of professionals to work harder to attract and support more candidates to become professionally qualified.

Robustness analysis
Other studies used the rule of law as a control variable as a proxy for coercive (Hillier et al., 2011;Martinez-Ferrero & Garcia-Sanchez, 2017). Both studies have used this control variable while analysing the firm-level data, which is not the case in our study as the firm-level data for ASEAN start to be available only in 2015. This prevents authors from conducting studies at the firm-level. However, the same model as in Equation (1) was running again using the new control variable. The results suggest the same trend, so our model proves stable when controlling for the law factor. 7 Simultaneously, we have added income level as a control variable for the degree of national governance (Ramanna & Stellen, 2009). We argue that the degree of national governance affects a country's decision to adopt GIPS. Following Lasmin (2011), we used the income level variable from the World Bank as a proxy for this variable. 8 Results followed the same trend and are available from authors, although they are not reported. Further combinations have been undertaken like running the model using the raw figures of CFA holders, none of earlier conclusions was affected; the normative proxy is still the highest. Foreign Direct Investment, net inflows (% of GDP) as an alternative proxy for economic pressure (Lasmin, 2011), among others. Again, the trend of results is not affected; thereby proving the robustness of the results.

Summary and conclusion
Based on the earlier sections, our findings can be summarised over the following two points: Based on the earlier sections, our findings can be summarised over the following two points: (1) the adoption rate of GIPS in the ASEAN region is at a moderate level, with 31 per cent in 2015, and 33% in 2016. Compared to the 40% adoption rate of GIPS currently in the EU (GIPS, n.d.), the development is reasonable, considering the gap of social and economic development between these two regions.
(2) Social pressures (measured by development assistance level, stock market maturity, and professional network) jointly have a greater impact than economic pressure (measured by GDP growth) in the adoption of GIPS. The findings partially support the results in Lasmin (2011) andMartinez-Ferrero andGarcia-Sanchez (2017), and confirm that the social pressures that affect the adoption of IFRS and Sustainability Standards also apply to the GIPS case. Within the social pressure components, normative dimension (CFA membership as the proxy) has the largest impact over the mimetic (MC as the proxy) and coercive (ODA as the proxy) dimensions. Coercive isomorphic change has the least influence for the adoption of GIPS.
As the first study on this topic in the ASEAN region, with rigorous research design and robust analysis, our paper contributes to the literature in three ways.
First, we initialised the research using an institutional theory framework in order to examine the influences on the adoption of voluntary standards for investment performance measurement by providing a clear picture about GIPS adoption in the ASEAN region-an overlooked area, from which, a distance can be identified with other regions (for example, the EU) that contain developed countries.
Second, our findings have identified both the most and least important components within social pressures with normative dimension as being more important than mimetic, while coercive is the least influential factor in the adoption of GIPS.
Third, we made a methodologic contribution by combining data from two different sources and presenting relevant results for current and future studies.
Our findings have important policy and managerial implications. On the one hand, policymakers in the ASEAN should focus on the reinforcement of legislation and regulation infrastructure in order to improve the political, economic, and cultural integration in the region as a whole, to support the implementation of international professional standards such as GIPS, and on the other hand, they should also recognise that the profession or professional networks can and have played a significant role in the promotion of these international standards, evidenced by our findings which indicate that normative isomorphic change has made a useful contribution to the adoption of GIPS.
The current study has limitations at regional level as a result of data availability. However, while an objective of the ASEAN is to become more integrated, it might be a good idea to replicate our study in a few years when adoption would be more widespread for non-adopting countries, and more established for the adopting countries. Moreover, it expects the professional (normative) dimension would remain dominant in the field, thus it is interesting to see the movement towards capital market dimension. Our study found that approximately 95 per cent of country-year observations in the adoption category have a capital market in their countries, and therefore we expect this variable to develop more influence in the future. 9 A firm-level study is needed once the data become available.
Furthermore, market-based research is recommended as we expect a more significant role in this dimension. Once this is proven, it would be excellent to see a capital market study in this field, once more data are available. Finally, we deem that it would be an excellent idea moving forward to integrate between country level and firm-level, once both studies are conducted separately. We regard our study as the base of the recommendations mentioned earlier, and this can be thought of as an additional contribution alongside our results. Lasmin, 2011;Martinez-Ferrero & Garcia-Sanchez, 2017). 5. Logistic Regression model the probability (P) of an even that is the Adoption of GIPS in our study. Thus, the extended form of the dependent variable is lnð p 1Àp Þ. 6. This problem occurs when there is a strong correlation between one or more of the independent variables. We will use two indicators as explained in the text to detect such a potential problem: Variance Inflation Factor (VIF) and tolerance statistics. 7. The same trend does not indicate same figures; significant level remains within the same ranges of the first outcomes, so conclusions about the hypotheses stay the same. Detailed computer outputs are available from authors upon request. 8. Income level in the World Bank database is four levels, ASEAN countries are distributed based on the countryyear observation as 58 (32.2%) low income, 61 (33.9%) lower-middle-income, 25 (13.9%) upper middle income, and 36 (20%) High-income economics. In the model, we created dummy variable two levels; zero for low level and one otherwise. 9. This figure arises from running cross tabs for adoption status against a dummy variable for the existence of the capital market. We are presenting this portion, as we believe it may be relevant for future studies, and we are happy to share any results, along with the data, with any researcher in the future.