Heterogeneous Patterns of Financial Development: Implications for Asian Financial Integration

ABSTRACT

This paper analyzes detailed differences in patterns of financial development across the major Asian economies, including three of the region's largest economies (China, Japan and South Korea), to understand how these differences might affect possibilities for greater regional financial integration. In particular, the paper argues that heterogeneous patterns of financial development, and not just differences in levels of financial development, may present an economic challenge to regional financial integration efforts, aside from possible political challenges. The paper provides background on the case for financial openness, Asian experiences with financial integration, and regional economic responses to external shocks. It also discusses policy options, including regulatory reform and coordination, and possible risk management policies and institutions, in the context of heterogeneous patterns of financial development.

KEY WORDS:

• Financial development
• financial integration
• Asia Pacific
• East Asia

JEL CLASSIFICATIONS:

• F30
• F55
• G10
• G20
• O16

Acknowledgements

This is a revised version of a paper presented at the Inaugural International Conference on Evolving Finance, Trade and Investment in Asia, held at the Centre on Asia and Globalisation, Lee Kuan Yew School of Public Policy, National University of Singapore, on 16–17 September 2015. We are grateful to conference participants, and particularly to Siu Fung Yiu, our discussant, for extremely valuable comments and suggestions. We are also grateful to the conference co-organizer and special issue editor, Tomoo Kikuchi, for valuable help. All remaining shortcomings are solely our responsibility.

Notes

1 On regional production networks, trade, and growth, see Athukorala (2014) and Kaur (2014), as two examples of a large and growing literature.

2 See Aizenman, Chinn and Ito (2013) as one of several articles by those authors that document and analyze this reserve accumulation.

3 Arguments for this view are taken up in the next section, where references are provided.

4 The main challenge for macroeconomic policy is encapsulated in the idea of the policy ‘trilemma,’ or ‘impossible trinity,’ based on the Mundell-Fleming model of an open economy macroeconomic framework. In the model, it is impossible for a government to simultaneously have monetary policy autonomy (and hence the ability to control the domestic inflation rate) and a fixed exchange rate when the capital account is completely unrestricted. Attempts to conduct an independent monetary policy will drive a wedge between foreign and domestic interest rates, leading to continued capital inflows or outflows (depending on the direction of the interest differential) in the absence of an equilibrating mechanism such as exchange rate adjustment. Rey (2013) has advanced the proposition that the weight of capital flows in the context of non-conventional monetary policies (essentially, what is known as QE or Quantitative Easing) makes exchange rate flexibility insufficient for domestic inflation control unless there are also controls on international capital flows. The theory and empirics of this view are still being debated. Aizenman, Chinn and Ito (2013) have been among the originators of quantitative approaches to measuring policy stances with respect to the trilemma.

5 For an academic statement of the changed thinking, see Ostry, Ghosh, Chamon, and Qureshi (2012). Ghosh et al. (2008) provide an earlier, more policy-focused take with the same perspective, also from the IMF.

6 Other prominent arguments against full capital account liberalization include Bhagwati (1998), Cooper (1999), Obstfeld (2009), and Stiglitz (2003).

7 Kaur and Singh (2014) provide a detailed review of the empirical evidence on how different East Asian economies reacted to the financial crises of 1997–1998 and 2007–2009, and the impacts of different policy mixes with respect to financial openness.

8 There is also evidence that the specific nature of capital flows matters. For example, equity flows have a positive short-run impact on the host economy (Henry, 2007; Kose, Prasad, & Terrones, 2009), as does foreign direct investment (e.g., Kose et al, 2009).

9 Arcand et al. (2015a) also test a piecewise linear model with exogenous thresholds of financial depth.

10 Law and Singh (2014) survey other possible explanations for the negative impact of financial depth on growth. Cline (2015) argues that all these results are merely a statistical artifact, and claims that the results are explained by slower growth at higher per capita income levels, but this claim seems to have been answered effectively by Arcand, Berkes, and Panizza (2015b).

11 The interpretation of this calculation is discussed in more detail later in this section.

12 More specifically, the term used is Financial Market Development, and it is one of 12 ‘pillars’ of the GCI.

13 Implicitly, this calculation treats all the components of the index as having equal weights. This is not an issue in our calculations, since the index we use weights components equally.

14 In other words, the components of the vector are not draws from a single distribution, since they are measures of different aspects of financial development.

15 The legal rights measure is actually included in the institutional environment component, but is only one of several institutional factors in that case.

16 We are indebted to our discussant, Siu Fung Yiu, for suggesting construction of these graphs.

17 In this context, the observations of the IMF (2015) report are of interest, ‘Differences in financial regulation between countries are important determinants of financial integration, as investors may be reluctant to carry out financial transactions with entities in countries whose regulations and institutions are very different from their own.’

18 We are again indebted to our discussant, Siu Fung Yiu, for suggesting construction of these figures.

19 Once more, we are indebted to our discussant, Siu Fung Yiu, for suggesting that we conduct a cluster analysis.

20 For cluster analysis based on trade intensity, see Huang, Huang and Sun (2006); for analysis based on FDI intensity, see Yang and Huang (2014).

21 For example, we discovered that the clustering was quite sensitive to the way in which the original measure of investor legal rights was converted to the 1–7 scale.

22 Vietnam, which is initially clustered with Korea in the levels analysis, later gets added to this cluster based on patterns. Sahay et al. (2015) also discuss the comparison of South Korea and Vietnam, where both countries have similar private credit but differ strongly along the financial access dimension.