Towards rule-based institutions and economic growth in Asia? Evidence from the Asian Financial Crisis 1997

Abstract

Due to the severity of the Asian Financial Crisis 1997–1998, South Korea, Indonesia, and Thailand resorted to an International Monetary Fund (IMF) bailout. In exchange, the IMF demanded a series of reforms intended to promote rule-based institutions generally found in advanced Western economies, such as the rule of law. Using panel data analysis from 1982 to 2007, we test empirically whether judicial independence, one of the more fundamental rule-based institutions, can positively explain the growth of these countries after the crisis, and find the impact of reforms to be limited. To understand why, we use South Korea as an example to show that top-down reforms by the government prevented a shift towards a rule-based economy. Due to the government selectively bailing out big businesses, big businesses that survived the crisis captured market shares once owned by the dissolved big businesses, becoming too powerful for the government to regulate. This research uses Soifer’s theoretical framework on critical junctures.

Keywords

Asian Financial Crisis critical juncture IMF reform institutional change

Introduction

Much research has examined different institutional mechanisms of economic growth in Asia and advanced Western states in a comparative manner. Today, existing comparative studies generally agree on two findings. First, Asia achieved tremendous growth, despite lacking rule-based institutions generally found in the advanced West such as protection of property rights, independent courts, and the rule of law. Second, in Asia, a strong government characterized by government–business networks substituted the functions of rule-based institutions (Haggard, 2004; Kohli, 2004; Wade, 1994).

One question that existing studies have yet to agree on an answer to is whether Asia will continue to rely on government–business networks for economic growth, or eventually transform and rely on rule-based institutions to achieve economic growth. Renowned economist Dixit (2009) posits that transformation towards rule-based institutions is inevitable. As economies grow and become more complex, they have no choice but to transform to manage the increasing transaction costs. In Asia’s context, the government–business networks would eventually have no choice but to change towards rule-based institutions, a system more suitable for managing complex transactions on a larger scale. In sum, for Dixit, rule-based institutions found in the advanced West are the linear progression of the government–business networks found in Asia.

JS Li (2003) and Rajan and Zingales (1998) use Dixit’s framework to explain the Asian Financial Crisis 1997–1998. For them, the crisis showed the limits of government–business networks – as the complexity of financial flows and investments made it difficult for the government to control and manipulate its comparative advantage. Based on Dixit’s analogy, transaction costs for dealing with foreign investors proved too high and thus exposed the limits of Asia’s government–business networks.

In this context, one interesting question that has not been examined with empirical rigor is whether the crisis served as a turning point for progressing Asia’s government–business networks towards rule-based institutions. Based on existing studies on critical juncture and historical institutionalism, an exogenous shock such as the Asian Financial Crisis can cause uncertainty and lead to institutional change (Capoccia and Kelemen, 2007; Steinmo and Thelen, 1992). Reviewing existing studies, we find that not much research has empirically examined this likelihood in Asia. Of course, regardless of how devastating an exogenous shock is, institutions are sticky due to inertia created from path dependency; in Asia, government–business networks have been the key driver of tremendous growth since the 1960s. But interestingly, since the late 1980s, South Korea, Indonesia, Thailand, Japan, and Malaysia were already pursuing a rapid financial liberalization, opening their capital accounts and gaining membership of the World Trade Organization in 1995 (Li, 2003).

These events indicate that before the crisis, constraints were already imposed on the government–business networks. On the surface, these constraints, combined with the Asian Financial Crisis 1997–1998 and the International Monetary Fund (IMF) intervention, may have transformed Asia’s government–business networks to rule-based institutions found in the advanced West. This expectation seems particularly applicable to South Korea, Indonesia, and Thailand, the three Asian states with no choice but to resort to the IMF for a bailout. In return, the IMF demanded significant reforms aimed at putting an end to the economy centered on government–business networks and promoting rule-based institutions such as the rule of law. Given that these three countries were pursuing a rapid liberalization, the IMF intervention and the shock of the crisis may have brought institutional change.

The aim of this research lies in testing whether such change came about in the three countries. Specifically, we want to examine whether the states that the IMF intervened in which have depended on a strong government unconstrained by law – government–business networks – for economic growth, have transformed towards a model of growth found in advanced Western states, i.e. rule-based institutions that constrain the government under the law after the crisis. To compare whether Asia’s strong governments unconstrained under law have become constrained under law, we use judicial independence. Effective judicial independence implies that the courts can successfully constrain the government under the law (Carrubba et al., 2015; Hayo and Voigt, 2007; Linzer and Staton, 2015). Thus, before the crisis, when Asia’s governments were unconstrained by law, we expect judicial independence in Asia to show a negative association with economic growth. After the crisis, if transformation of institutions took place, then we expect judicial independence in Asia to display a positive association with economic growth.

For empirical analysis, we conduct a fixed effect panel data analysis consisting of the three IMF intervened states – South Korea, Indonesia, and Thailand – and 21 advanced Western states from 1982 to 2007. We include 21 advanced Western states as a control group to test whether their rule-based institutions are a linear progression of the three IMF intervened states’ institutions. To show robust results, we reanalyze our results replacing our explanatory variable, judicial independence, with a similar variable, constraint on the executive from Polity4.

As our theoretical framework, we will employ Soifer’s (2012) approach of critical junctures that divide critical juncture into permissive condition and productive condition for explicit causal analysis. For an institution to diverge, or shift towards rule-based institutions, both permissive and productive conditions must be satisfied. A permissive condition would qualify as a necessary condition, and a productive condition as a sufficient condition for shift. Based on this framework, our analysis shows that the crisis did not have much impact. Before the crisis, judicial independence had a negative effect on economic growth in the three IMF intervened states, and after the crisis the effect does not change. In sum, the Asian Financial Crisis 1997–1998 satisfied as a permissive condition, but not as a productive condition.

To elaborate why path dependency prevailed, we use South Korea’s case to show that the incumbent president Kim Dae Jung’s reform initiative, the ‘Big Deal’, was aligned with the existing model of government intervention. Thus, Korean government’s top-down reforms selectively saved big businesses based on political decisions, rather than on the merits of the market. Such reform brought detrimental consequences, for big businesses that survived the crisis captured the market share of the big businesses that failed to receive bailouts. As a result, big businesses that survived the crisis became too powerful for the government to regulate.

Institutions, the Asian Financial Crisis, and IMF reform

This section begins with a brief comparison between institutions in Asia and in the advanced West; in the advanced West, institutions have been centered on constraining the government under the law, whereas in Asia they have been centered on a strong government unconstrained under the law. In the next section, we briefly discuss the causes of the Asian Financial Crisis 1997–1998, followed by the conditions the IMF required in exchange for bailing out the three Asian states – South Korea, Indonesia, and Thailand. In the final section, we introduce our theoretical framework based on Soifer’s (2013) causal logic of critical junctures. Using this framework, we hypothesize that conditions of critical juncture did not fulfill their stated purpose, and thus the IMF intervened states did not experience reform towards rule-based institution.

Different mechanisms of economic growth in Asia and the advanced West

Much research has compared the different institutional mechanisms found in Asia and in the advanced Western states. For the advanced Western states, the notion of the institution centers on DC North’s (1990, 1991) definition, a “rule of the game” or set of rules that constrain the behavior of actors such as the state or a private agent. Such constraints resolve around collective action and credible commitment problems by lowering transaction costs. As a result, they induce market activities such as investment, trade, and labor activities, and lead to economic growth. Examples of such an institution are the rule of law, the security of property rights, contract enforcements, and independent courts, also identified as rule-based institutions that constrain actors under law to prevent acting on their whims, thus leading to growth (Acemoglu et al., 2005; Glaeser et al., 2004; Haggard et al., 2008). For this research, we will label institutions found in the advanced West as rule-based institutions. Synonymous expressions found in other studies are institutional quality, formal institutions, and rule-based governance, with some scholars identifying them as institutions found in the advanced West (Levchenko, 2007; Torgler and Schneider, 2007; Williamson, 2009; Williamson and Kerekes, 2011).

On the other hand, influential research on institutional mechanisms of Asia suggests a contrasting notion of institutions from the advanced West. Dubbed East Asian Miracle states¹ – South Korea, Japan, Singapore, Indonesia, Thailand, Malaysia, Taiwan, and Hong Kong – these states have achieved tremendous economic growth from the 1960s to the early 1990s, despite lacking effective rule-based institutions. Rather, the role of the government played a key role (Evans and Rauch, 1999; White and Wade, 1988), as it intervened to substitute the functions of the rule of law, such as credible commitment and collective action problems (Haggard, 2004). While much literature has examined the role of the government in Asia, it commonly emphasizes the government–business networks and the counsel, which lowered transaction costs to induce economic activities and prevent predation (Haggard, 2004; Kohli, 2004). In sum, Asia promoted a planned economy which enabled it to channel resources more effectively to promote industrial policies, enabling it to manipulate its competitive advantage to compete against the world rather than accepting the natural comparative advantage (Amsden, 1994; Leipziger, 2001). Such an approach contrasted from that of the advanced West, which relied on forces of market economies with limited government intervention. For the purpose of this research, we will identify these as informal institutions, which includes definitions such as government–business networks and trust-based institutions (Hans-Joachim, 2015; Helmke and Levitsky, 2004; Li, 2003).

Today, existing studies generally accept that various modes of paths exist for economic development (Acemoglu et al., 2008; Li, 2013). But these works lack consensus as to whether informal institutions will eventually converge towards rule-based institutions or remain divergent. Scholars such as Kurtz and Schrank (2007) acknowledge that developing countries may lack advanced rule-based institutions and thus may rely on informal institutions for development. But it is unclear whether developing countries would eventually transform towards rule-based institutions, especially for Asia, where informal institutions – government–business networks – have spanned the economy for so long. Some scholars have argued that due to the culture of a trust-based society being embedded in Asia, most notably in China, a transition towards a rule-based economy will be limited (Li, 2013, 2014). On the other hand, renowned economist Dixit (2009) maintains that transformation is inevitable, as countries become more reliant on strangers for business, which requires a system based on law and impersonality rather than trust to deal with rising transaction costs.

Causes of the Asian Financial Crisis

Much research has examined the causes of the Asian Financial Crisis 1997–1998. According to Li (2003), the crisis took place because the government’s capacity to monitor complex financial flows and foreign investments was weakened before the crisis, when many East Asian states attempted rapid political and financial liberalization. To make his case, Li identifies a common trend found in the IMF intervened states – South Korea, Indonesia, and Thailand – before the crisis. After tremendous growth in the 1980s and 1990s, South Korea, Indonesia, Thailand, Malaysia, and Japan focused on achieving rapid financial liberalization, beginning with a drastic opening of their capital accounts. At the same time, Korea and Thailand were also spearheading political liberalization. South Korea wanted accession to the Organisation for Economic Co-operation and Development (OECD) and an accelerated financial liberalization process in 1993, while at the same time, political liberalization brought competitive elections and also the dismantlement of the Economic Planning Board (EPB) which has served as a control tower of the economy since the 1960s (Jung, 2011). Thailand adopted a de jure competitive electoral system in 1992, which made the government extremely unstable, with a new government coming to a power every year (Li, 2003). Due to the liberalizations, trust and networks between the government, financial institutions such as banks, and the big businesses weakened, as they now had to abide by international standards. As a result, the government could not effectively monitor the complexity of financial flows.

Conditionality for the IMF’s bailout reforms

Due to the severity of the crisis, South Korea, Indonesia, and Thailand had no choice but to resort a IMF bailout. In exchange for bailing out South Korea, Thailand, and Indonesia, the IMF requested these countries to make structural reforms to their economies, briefly summarized as follows. Firstly, it demanded that these states develop a rule of law. In the past, these three states, rather than closing troubled financial institutions and businesses, continued to bail them out, leading to a moral hazard behavior between big businesses and the government (Lee, 2000; Yanagimachi, 2004). As Krugman (1998) points out, excessive risk-taking by the big businesses, knowing that the government would bail them out in case of bankruptcy, caused the crisis. Second, the IMF required improvement in accounting and disclosure rules since, as Li (2003) points out, a lack of transparency and a discrepancy between local and foreign information played a role in spreading the crisis. Third, the IMF demanded elimination of unfair business trade, from eliminating trade-related subsidies to promoting labor market flexibility. Fourth, it required the liberalization of property markets, thus enabling foreign ownership by completely opening financial and real estate markets. This policy comes from previous concerns that big businesses such as Chaebol in Korea were becoming too big to fail and, due to their ownership structure, had monopolistic control of the market. Fifth, the IMF required improvement in the quality of regulation by establishing a supervisory institution, the Financial Supervisory Service (FSS), which oversaw financial activities and established greater central bank independence. Furthermore, the IMF demanded change in corporate governance and enhanced regulation of corporations to end the monopolistic behavior of big businesses and their political ties to the government usually in the form of family connections (Freedman, 2005; Lee, 2008). Such ties were especially prevalent in South Korea, with Chaebols receiving a great deal of blame for the Asian Financial Crisis 1997–1998 due to incompetent family management (Lee, 2008). Overall, the IMF requirements aimed to transform relation-based Asian economies centering on state-led interventions towards market-based economies.

Institutional change, critical juncture, and the Asian Financial Crisis

The Asian Financial Crisis exposed limitations in Asia’s trust-based system. With a flow of rapid foreign investments and the rise of transaction costs for monitoring economic activities, the system could no longer be sustained. But does such a shock guarantee an automatic reform towards rule-based institutions? To answer this question, we need to examine existing theories that can explain institutional change in the context of path dependency. One of the commonly used approaches is critical juncture analysis, in which an exogenous shock can generate a period of influx and uncertainty large enough to cause a shift in the existing equilibrium (Capoccia, 2015; Capoccia and Kelemen, 2007). By definition, divergence from path dependency happens due to a loosening of existing constraints within the structure (Soifer, 2012).

To analyze whether the Asian Financial Crisis 1997–1998 qualifies as a critical juncture, this research will employ Soifer’s (2012) analytical approach. Compared to previous theories of critical juncture, Soifer’s approach is analytically useful, since it introduces an explicit causal logic for how critical junctures can lead to divergence from path dependency. Previous works faced limitations, as they did not specify the conditions for critical juncture. To achieve divergence, Soifer distinguishes two types of causal conditions: permissive conditions and productive conditions. Permissive conditions mean that constraints within the structure are loosened, thus making divergence from the existing path possible. However, the permissive condition itself does not guarantee a shift from path dependence. For divergence to take place, productive conditions also must be fulfilled. Productive conditions, within the confines of permissive conditions, are defined as producing outcomes that are reproduced when permissive conditions stop existing and thus critical juncture closes. In another words, it is “the aspects of a critical juncture that shape the initial outcomes that diverge across cases” (Soifer, 2012: 1574). In sum, productive conditions combined with permissive conditions to establish a condition necessary and sufficient for divergence.

Soifer (2012) provides an example of a permissive condition through the economic crisis in Latin America during the 1980s. Citing Weyland (2004), Soifer identifies the onset of hyperinflation as a permissive condition. As for productive conditions, Soifer (2012) lists two conditions as an example in the context of Latin America. First, the type of leadership, and second, the severity of the crisis. By type of leadership, Soifer means that radical shifts in economic policies are possible only if the leader is not committed to another developmental model. In Latin America, a radical shift in policies materialized only for the countries with a greater severity of the crisis. Together, two conditions served as the exhaustion of the existing developmental model. In sum, divergence materialized because hyperinflation served as a permissive condition, and the exhaustion of existing development served as productive conditions.

In the context of the Asian Financial Crisis 1997–1998, permissive conditions are the onset of foreign investors pulling out their investments and the weakening of currencies of respective intervened states are sets of the permissive condition. Through these events, existing structural constraints that sustained government–business networks were loosened. For example, in all three IMF intervened states, political change took place. In South Korea, a reform-minded opposition party won the presidency for the first time; in Indonesia, Suharto, who has been in the office for more than 30 years, was forced to resign; in Thailand, the incumbent prime minister also stepped down. The productive conditions, however, are more complicated. For example, in the case of South Korea, the new opposition party leader with a reform-minded president Kim Dae Jung came into power. However, the president faced complex choices, due to the IMF intervention. Like the IMF, President Kim wanted reforms. But, if President Kim accepted all demands, such as ending the moral hazard of the government bailing out bankrupted businesses to promote the rule of law, the majority of Korean businesses would be owned by foreigners, since the IMF also required the complete liberalization of property markets.

In this context, one could argue that although the permissive condition loosened existing structural constraints, in the process, it invited another structural constraint, the IMF intervention. Therefore, President Kim could not completely abandon the existing development model, since doing so would be politically unpopular. As a result, President Kim’s reform policy, the so-called ‘Big Deal’ with big businesses, could not completely abandon the existing development model of the government–business partnership. Thus, rather than letting reforms take place based on the merits of the market, the government intervened and made political decisions to selectively bail out big businesses. Although 25 big businesses in Korea were bankrupted (Feenstra et al., 2002), some such as Hyundai became even more powerful, as bankrupted automobile company Kia became Hyundai’s. Existing studies cite the Big Deal as a disappointment and continuation of the existing developmental model based on the network of government and big businesses (Cherry, 2005). In sum, we argue that the productive condition was not met due to the constraints on the reform-minded President Kim. Our hypothesis is consistent with existing arguments that identify the nature of path dependency between government–business relations as difficult to reform (Lee, 2008; Park, 2011), and also with North (1990) who maintains that reform attempts may fail due to political obstacles. Similarly, our argument also follows Dixit (2009), who maintains that a top-down reform by the government generally has limited success, because it lacks the incentive mechanisms and local knowledge. In the next section, we introduce our empirical strategy to test our prediction.

Empirical strategy

Our empirical analysis centers on two-way fixed effects panel data analysis of IMF intervened and advanced Western states from 1982 to 2007. Rather than rely on a single panel data analysis, we conduct multiple panel data analysis to validate our results, which we further explain in the section under results. But first, we explain our empirical model, including how to measure informal institutions and what qualifies as reform towards rule-based institutions. Then, we explain specification of our empirical model, followed by the data, describing our model’s dependent, independent, and control variables.

Empirical model

This section begins with a question of how to empirically capture whether reform has been successful. Based on existing literature, we learned that before the Asian Financial Crisis, the three Asian states (South Korea, Thailand, and Indonesia) relied on informal institutions – network and trust between big businesses and the government – to achieve economic growth. Since the IMF reforms centered on promoting rule-based institutions such as the rule of law and limited government intervention found in the advanced West, successful reform implies rule-based institutions can explain growth in the three states. More fundamentally, this would imply that, as in the advanced West, in Asia constraining the economic actors – government and big businesses – under the law rather than strong government interventions can resolve credible commitment problems and foster economic activities.

To measure rule-based institutions and the rule of law, we use judicial independence. Consistent with the existing literature, a country with an effective independent judiciary means it can constrain the executive branch under the law (Linzer and Staton, 2015; Weingast, 1995), a condition found under advanced Western states. On the contrary, when the judiciary is not independent, it will make political decisions, which thus implies that it cannot constrain the executive branch under the law, a condition found in judicial courts in the East Asian Miracle states (Li, 2003). Since a government unconstrained by law signifies a low level of judicial independence, we expect judicial independence to be negatively associated with economic growth for the three IMF intervened states before the crisis; whereas the advanced Western states will display a positive and statistically significant coefficient. To test whether the IMF intervened states are transforming towards rule-based institutions, the advanced Western states – 21 countries from the EU, the US, and Canada – are included as a reference group. These countries are chosen as a base group for comparison because they have barely been affected by the crisis and represent economies of rule-based institutions (Haggard, 2004; Li, 2003).

Model specification

We begin with a specification of our empirical model. To investigate institutional effects on economic growth, this research will conduct several panel data analyses, both to test the degree of change after the Asian Financial Crisis and to conduct a robustness check. But first, we employ fixed effects panel data analysis with both country and year fixed effects. The base model is as follows:

Y_{ct} = β_{0} + β_{1} \cdot {JI}_{ct} + Z_{ct} \cdot Γ + τ_{t} + η_{t} + ε_{ct}

Where JI _ct is the measure of judicial independence for a given country c in given year t, vector Z_ct · Γ represents control variables for a given country c in given year t, and τ_t represents year fixed effects and η_t country fixed effects. Ultimately, we want to compare the effects of the Asian Financial Crisis on the IMF intervened states with reference to the advanced Western states. Below is the model specification with multiple interaction variables:

\begin{array}{l} Y_{ct} = β_{0} + β_{1} {JI}_{ct} + β_{2} {JI}_{ct} \cdot G_{IMF3} + β_{3} {JI}_{ct} \cdot {AfterCrisis}_{t} + β_{4} G_{IMF3} \cdot {AfterCrisis}_{t} \\ + β_{5} {JI}_{ct} \cdot G_{IMF3} + \cdot {AfterCrisis}_{t} + β_{6} {AfterCrisis}_{t} + Z_{ct} \cdot Γ + τ_{t} + η_{t} + ε_{ct} \end{array}

Where G_IMF3 represents a country group dummy variable for the IMF intervened states, South Korea, Indonesia, and Thailand, and the reference group (when IMF3 = 0) equals advanced Western states consisting of EU member states, the United States, and Canada. The variable AfterCrisis_t represents the years after the Asian Financial Crisis – from 1999 to 2007 – and the reference year, or when AfterCrisis_t = 0, represents the years before the crisis, i.e. 1982 to 1996.

Data

Our dependent variable is GDP growth rates from the World Development Indicator. Figure 1 shows GDP growth rates of the IMF intervened states, advanced Western states, and the rest of the world. Although we do not include the rest of the world in our analysis, we have included it as a reference along with advanced Western states to illustrate its different trend with IMF intervened states. As shown in Figure 1, the fluctuation in growth rates differs between the IMF intervened states and the other two groups.

Figure 1.

Average economic growth rates, 1982–2007.

Before the crisis, the IMF intervened states, represented in color green, achieved a much higher growth rate than the other two groups. During the crisis, the IMF intervened states suffered one of the worst economic recessions of their history, evidenced by the steep decline in 1997–1998. For the other two groups, the crisis did not have much effect. After the crisis, however, the three groups achieve a growth rate at a similar level, with a smaller gap compared to the period before the crisis.

For the explanatory variable, judicial independence, we use the index developed by Linzer and Staton (2015). A composite index ranging from 0 to 1, this measure examines how much power or influence the judiciary has compared with other branches such as the executive and legislative branches. Figure 2 shows the trend from 1982 to 2007 for the IMF intervened states, advanced Western states, and the rest of the world.

Figure 2.

Mean of judicial independence, 1982–2007.

Unsurprisingly, advanced Western states remain at the top. Both the IMF intervened states and the rest of the world gradually improve, and eventually the IMF intervened states supersede rest of the world beginning in 1991. This trend is consistent with the findings of Li (2003) that the IMF intervened states attempted political and financial liberalizations during the late 1980s and early 1990s. Despite substantial improvements, however, a significant gap between the IMF intervened states and the advanced Western states remains throughout the crisis, and further diverges in 2007.

For control variables, we include variables that the existing literature identifies as pertinent for growth, also available from 1982 to 2007. These variables are government consumption (% of GDP), trade (% of GDP), population (log), gross capital formation (% of GDP), gross savings (% of GDP), and GDP. Trade, often used as a measure of market integration and openness of the economy (Rodrik et al., 2004), has often been linked with growth. Labor and capital, two components of growth identified from the endogenous growth theory (Aghion et al., 1998; Krueger, 1995), are perhaps the oldest determinants for growth. For this article, we use population as a proxy for labor and gross capital formation for physical capital. Other determinants of growth, gross savings and government consumption are also included in the model, while controlling for the GDP of each country. For all control variables, we used the World Development Indicator (WDI).

Results

To verify our empirical results with robustness, multiple regression analyses are conducted. Our empirical analysis begins with fixed effects panel data analysis, which confirms our theoretical expectation that judicial independence positively affects growth in the advanced Western states, but negatively affects growth in the IMF intervened states, with not much change taking place after the Asian Financial Crisis 1997–1998. To confirm our initial findings, lead years analysis is also conducted. Since our results may potentially be affected by reverse causality and path dependency, we also employ dynamic panel data estimation using GMM, which confirms our initial results from fixed effects analysis. Finally, as a robustness check, we substitute our explanatory variable with a variable of a similar concept, constraint on the executive from Polity IV. The results are consistent with what we obtained from the initial explanatory variable. In sum, our results, after rigorous verification, show that the crisis had a limited effect on transformation towards rule-based governance in Asia.

Main findings

Table 1 shows the estimation results for how judicial independence affects economic growth for the three IMF intervened states with reference to the advanced Western states. For control variables, gross capital has a positive and statistically robust effect on growth across all models. Trade and savings also have a positive effect, but only for the OLS models.

Table 1.

Effects of judicial independence on economic growth.

	No controls		With controls		With interaction of treatment group
	OLS (1)	FE (2)	OLS (3)	FE (4)	OLS (5)	FE (6)
JI	-4.437***	-3.697	-1.766	-1.275	1.241	5.565***
	(0.804)	(2.809)	(1.180)	(3.629)	(0.989)	(1.888)
Trade			0.00931***	0.00766	0.0102***	0.00666
			(0.00310)	(0.0172)	(0.00299)	(0.0207)
Gross capital			0.235***	0.246***	0.241***	0.263***
			(0.0454)	(0.0448)	(0.0465)	(0.0369)
Population			0.253	-1.742	-0.286	2.548
			(0.312)	(4.640)	(0.328)	(3.731)
Gross savings			0.0149	0.0420	-0.0144	0.111**
			(0.0254)	(0.0748)	(0.0225)	(0.0517)
Gov. consumption			-0.00833	0.0108	0.0142	-0.00175
			(0.0325)	(0.173)	(0.0314)	(0.136)
GDP			-0.245	-1.676	0.211	-0.880
			(0.288)	(1.335)	(0.298)	(0.788)
IMF3					6.438***
					(1.445)
JI * IMF3					-6.619***	-20.37***
					(2.317)	(4.121)
Constant	5.088***	4.495*	-1.970	64.75	-7.548**	-31.18
	(0.826)	(2.359)	(3.004)	(77.72)	(3.135)	(56.75)
Controls	No	No	Yes	Yes	Yes	Yes
Country fixed effect	No	Yes	No	Yes	No	Yes
Year fixed effect	Yes	Yes	Yes	Yes	Yes	Yes
Observations	624	624	540	540	540	540
R-squared	0.223	0.191	0.392	0.295	0.428	0.368
Number of countries		24		24		24

Notes: ***p < 0.01, **p < 0.05, *p < 0.1; robust standard errors are used.

IMF3, or the three IMF intervened states, includes South Korea, Indonesia, and Thailand. Reference group, or when IMF3 = 0, countries are advanced Western states.

Other control variables – GDP, population, and government consumption – are not statistically significant.

The explanatory variable, judicial independence, exerts a negative effect on growth under OLS in model 1 but exerts positive effects under Fixed Effects in model 6. It should be noted that the coefficients for the two models carry different meanings. In model 1, the coefficient is an average of both the three IMF intervened states and the advanced Western states, treatment and reference groups we theorized to share contrasting characteristics. On the other hand, in model 6, the coefficient means the effect of judicial independence on growth only for the advanced Western states, or the reference group (IMF3 = 0), since the coefficient of the interaction variable JI * IMF3 measures the marginal effects of judicial independence for the IMF3 group compared to the reference group. In this regard, the results in model 6 confirm our theoretical expectation, as judicial independence is positive and statistically significant with coefficient at 5.565***. For the interaction variable, JI * IMF3, the coefficient -20.37*** means that for the three IMF intervened states, the overall effect of judicial independence is -14.805.

Since Table 1 does not incorporate effects of the Asian Financial Crisis, we now turn to the Table 2. Each model, from 1 to 7, represents periods after the Asian Financial Crisis. Here, we are interested in gauging the degree of reforms after the crisis. If productive conditions were satisfied, we expect these states to achieve growth despite the government being constrained to a greater degree under the law than before the crisis. For our analysis, we leave out the years the crisis caused a sharp drop in the economy, 1997 and 1998, to better capture the effects of reform. Since the crisis lasted from 1997 to 1998, each period starts from 1999 and adds a year for the next period. The intention here is to examine the duration of the reform that took place after the shock, including the reform that IMF intervened countries undertook in exchange for the bailout.

Table 2.

Effects of the Asian Financial Crisis on judicial independence and economic growth in different periods.

After Crisis:	1999–2000 (1)	1999–2001 (2)	1999–2002 (3)	1999–2003 (4)	1999–2004 (5)	1999–2005 (6)	1999–2006 (7)	1999–2007 (8)
JI	6.564**	5.928*	6.957**	6.968**	6.900**	7.817***	7.803***	7.390**
	(2.921)	(2.925)	(2.758)	(2.844)	(2.948)	(2.781)	(2.759)	(2.742)
Trade	0.0284	0.0309	0.0253	0.0266	0.0184	-0.00139	-0.00429	-0.00289
	(0.0276)	(0.0233)	(0.0216)	(0.0216)	(0.0208)	(0.0209)	(0.0211)	(0.0218)
Gross capital	0.376***	0.362***	0.347***	0.347***	0.349***	0.340***	0.317***	0.309***
	(0.0920)	(0.0805)	(0.0807)	(0.0783)	(0.0745)	(0.0643)	(0.0539)	(0.0559)
Population	-0.441	0.791	1.462	2.822	3.335	3.297	1.425	1.604
	(4.227)	(4.062)	(3.572)	(3.483)	(3.496)	(3.224)	(3.121)	(3.422)
Gross savings	0.257***	0.252***	0.236***	0.238***	0.223***	0.198***	0.177***	0.150***
	(0.0587)	(0.0541)	(0.0523)	(0.0545)	(0.0541)	(0.0499)	(0.0470)	(0.0457)
Gov. consumption	0.100	0.103	0.0594	0.0805	0.0858	0.0383	0.00503	-0.00380
	(0.208)	(0.189)	(0.163)	(0.154)	(0.153)	(0.152)	(0.158)	(0.171)
GDP	-0.0629	0.0459	0.118	0.0713	-0.164	-0.659	-0.858	-0.849
	(1.145)	(1.067)	(1.025)	(0.940)	(0.916)	(0.926)	(1.032)	(1.106)
JI * IMF3	-41.88***	-41.06***	-40.87***	-41.04***	-39.49***	-36.45***	-32.60***	-29.56***
	(5.177)	(5.233)	(5.232)	(5.114)	(4.517)	(4.239)	(4.256)	(4.150)
Year	6.766	7.735	8.044	8.215	7.899	8.141	8.169	8.351*
	(5.124)	(5.537)	(5.592)	(5.846)	(5.443)	(4.940)	(5.352)	(4.803)
JI * Year	-3.825	-5.002	-5.590	-5.910	-5.157	-4.577	-5.051	-4.497
	(5.040)	(5.416)	(5.520)	(5.811)	(5.463)	(4.924)	(5.083)	(4.785)
IMF3 * Year	5.322	4.165	3.894	4.997	6.056	4.893	2.683	2.012
	(6.741)	(6.884)	(6.789)	(6.921)	(6.708)	(6.982)	(7.664)	(7.783)
JI * IMF3 * Year	-1.647	-1.636	-0.874	-3.054	-5.019	-3.664	-1.252	-0.718
	(8.148)	(8.576)	(8.216)	(8.578)	(8.886)	(10.03)	(11.03)	(11.39)
Constant	-12.35	-34.44	-46.11	-67.59	-69.33	-54.21	-17.05	-19.08
	(71.61)	(65.67)	(57.33)	(54.63)	(53.77)	(49.11)	(48.79)	(52.89)
County fixed effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Year fixed effect	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Observations	381	402	424	446	468	492	516	540
R-squared	0.486	0.483	0.473	0.474	0.463	0.445	0.425	0.403
Number of countries	21	21	22	22	22	24	24	24

Notes: ***p < 0.01, **p < 0.05, *p < 0.1; robust standard errors are used.

Thus, each period can be considered a treatment period, or years in which the effects of reforms lasted. Each treatment period is coded 1, whereas the base period, years before the crisis, or from 1982 to 1996, is coded 0. Consistent with the results from Table 1, for all periods before the crisis, judicial independence exerts a positive and statistically significant effect on growth for advanced Western states, evidenced by the positive coefficient of JI. For the IMF intervened states, the coefficient of JI * IMF3 is negative and statistically significant, implying that judicial independence exerts a negative and statistically significant effect on growth.

After the crisis, we find similar results across all periods for both the advanced Western states and the three IMF intervened states. For the advanced Western states, the effects of judicial independence after the crisis are not different from those in the years before the crisis, as JI * Year is not statistically significant across all periods. Similarly, for the IMF intervened states, the effects of judicial independence after the crisis are not different from in the years before the crisis, as JI * IMF3 * Year is not statistically significant across all periods.

Robustness check

In this section, we examine whether our results from the previous section hold under additional empirical rigor. Specifically, one may raise a concern with subjective measures such as judicial independence. Thus, we reanalyze our main results replacing the explanatory variable judicial independence with a similar index. For this purpose, we use the constraint on the executive developed by Polity4. We particularly use this index because it is widely used in the existing literature and is one of the few indexes with years available from the early 1980s. Conceptually, this index measures how many constraints are imposed on the executive, which slightly differs from the measure of judicial independence we used, which measures the power and influence of the judiciary, including influence on the executive but also on the legislative and on society. Since they share many similarities, we reanalyze our main results with this index. Table 3 shows the results of this.

Table 3.

Fixed effects panel data analysis.

	Model 1	Model 2
Xconst	1.313***	1.474***
	(0.392)	(0.352)
Xconst * IMF3	-1.971***	-2.670***
	(0.409)	(0.353)
Xconst * After Crisis		1.020**
		(0.407)
IMF3 * After Crisis		0.890
		(3.955)
Xconst * IMF3 * After Crisis		0.466
		(0.542)
Constant	9.583	18.39
	(60.02)	(72.89)
Country fixed effect	Yes	Yes
Year fixed effect	Yes	Yes
Controls	Yes	Yes
Observations	479	479
R-squared	0.340	0.373
Number of country	22	22

Notes: ***p < 0.01, **p < 0.05, *p < 0.1; robust standard errors are used.

The results are consistent with our previous results from fixed effects panel data analysis. The variable Xconst., effects of executive constraint for advanced Western states, is positive and statistically significant, while Xconst * IMF3, effects of executive constraint for the three IMF intervened states before the crisis, is negative and statistically significant, and Xconst * IMF3 * After Crisis, effects of executive constraint on growth for the three IMF intervened states after the crisis, is not statistically significant.

Overall, empirical analysis confirmed our projection that despite establishing a permissive condition, critical juncture could not lead to divergence due to not fulfilling the productive condition. Specifically, our analysis shows that the reform towards a rule-based economy has been limited for the three IMF intervened states.

Exploring the mechanism for lack of reform

In this section, we elaborate on why selective bailouts by the Korean president destroyed the remote possibility of reform towards rule-based institutions after the crisis. As a result of the selective bailouts, big businesses became too powerful to regulate after the crisis. One example that epitomizes the strength of big businesses in Korea after the crisis is their ability to defend against reform-minded presidents. After the crisis, the liberal party president Kim Dae Jung pledged to regulate chaebols as demanded by the IMF and introduced several policies reforming ownership structure and corporate governance (Lee, 2000). One of the key reform agendas was lowering indirect cross ownership, which allowed the Chaebols to retain their control and sustain their monopolistic nature (Choi and Kang, 2014; Lee, 2000). Similarly, President Roh, the president after Kim Dae Jung, also pledged to oblige by similar reform agendas. Their failure is displayed in Table 4, which shows the indirect cross ownership (%) for the top 10 chaebols in Korea before and after the crisis.

Table 4.

Indirect cross ownership (%) for the top 10 chaebols in Korea.

1992	1995	1998	2001	2004	2007	2010	2011
47.8	43.4	45.1	46.4	47.1	47.9	47.4	53.5

Source: Fair Trade Commission ( http://www.ftc.go.kr/eng/index.do).

Table 4 indicates that Korea’s liberal president’s reform agenda for chaebols has not been successful, as indirect cross ownership for the top 10 big businesses in Korea before and after the crisis has not changed much. This finding is consistent with existing studies, which consider chaebol reform after the crisis to be limited (Lee, 2008). If big businesses can retain indirect cross ownership, they do not need to worry about hostile takeovers and losing control of their businesses from foreign investors. Thus, despite Korea’s property markets being completely liberalized, this works to their advantage, as chaebols would not have to worry about foreign takeovers, but, simultaneously, can reap foreign investments without giving away their control. Combined with liberalization of the property markets, dissolving numerous chaebols also paved the way for the survived chaebols to increase their market share. During the crisis, 25 chaebols were bankrupted (Feenstra et al., 2002). In the case of South Korea, two liberal presidents with priority for reforming chaebols were in office since the crisis, Kim Dae Jung from 1998 to 2003 and Roh Moo Hyun from 2003 to 2008. President Roh promised to reduce the dependence on chaebols for the Korean economy, and to eradicate the government–business collusion and networks that had become the norm in Korea (Kihl, 2005). Yet, chaebol reform has been limited, as indirect cross ownership was higher in 2007 than in 1995.

In this context, we can conclude that informal institutions of trust between businesses and Korea remain after the crisis but in a different style compared to before the crisis. By the early 1990s, big businesses were becoming powerful enough to influence politicians and cut ties with the government (Okabe, 2015; Pempel et al., 2015). After the crisis, we expect this pattern of informal institutions to persist and continue. In sum, before the Asian Financial Crisis 1997–1998, the big businesses and the government had a partnership, with the government holding the clout due to its control of resources (Woo and Woo-Cumings, 1991). After the crisis, it is more likely that big businesses hold the necessary clout and resources to influence politicians to protect their interests.

Conclusion

This research has tested empirically whether Asia’s economy, which is reliant on informal institutions for growth, has transformed towards rule-based institutions after the Asian Financial Crisis 1997–1998.

Our panel data analysis with numerous robustness checks shows that the crisis had a limited effect on reform. To explain why, we used Soifer’s (2012) framework on critical juncture. Based on this framework, the Asian Financial Crisis 1997–1998 served as a permissive condition for divergence, but it was not enough, as productive conditions were not satisfied. Specifically, during the reform process, Korea’s President Kim decided to reuse the existing developmental model, rather than use a new model, due to the fear of political retribution. This decision brought lasting consequences, as it strengthened big businesses in Korea more than ever, and thus it was difficult to regulate them and to conduct reform.

This research makes several contributions to the existing literature. Firstly, it tests empirically whether Asia’s economies are path-dependent on its informal institution of government–business networks for economic growth. While various arguments have been made by the existing literature, not much empirical testing has taken place. Our empirical analysis shows that path dependency remains, as rule-based institutions such as judicial independence have limitations explaining economic growth in the three IMF intervened states after the crisis.

Second, this research makes a contribution to the existing literature by analyzing why path dependency prevails over reform. While much research has been done on institutional change and path dependency, we show that partial reforms may strengthen path dependency, as has been the case in South Korea. Despite the exogenous shock that brought an influx of uncertainty, the top-down reforms taken by the Korean government had a limited effect in reducing the influence of big businesses, thus making reform towards rule-based institutions difficult. In this regard, another contribution this research makes is applying the causal logic of critical junctures (Soifer, 2012). While critical junctures have been frequently used for explaining institutional change, not much research has applied the causal logic of critical junctures, which is more parsimonious and carries greater explanatory power than previous theories.

Finally, this research makes a contribution to the existing literature on the debate surrounding institutions in the advanced West. Scholars such as Dixit (2009) maintain that as economies grow and become more complex, informal institutions eventually have to transform towards rule-based institutions. Using panel data analysis from 1982 to 2007, we find that in Asia this has not been the case.

Of course, more work can be done to improve this research. For the future, further research can be done to enhance understanding of the scholarly community in two areas. Firstly, more analysis can be done to incorporate the global financial crisis in 2008 to test whether our findings in this research hold after the global financial crisis. Secondly, much of the analysis done in this research has focused on South Korea, one of the three IMF intervened countries. In the future, more research can be done on Indonesia and Thailand to better incorporate their national contexts.

Footnotes

Author’s note

Kee Hoon Chung is also affiliated with KDI School of Public Policy & Management, Sejong-si, Seoul, South Korea.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Kee Hoon Chung

Note

References

Acemoglu

Johnson

Robinson

(2005) Institutions as a fundamental cause of long-run growth. Handbook of Economic Growth 1: 385–472.

Acemoglu

North

Rodrik

, et al. (2008) Governance, growth, and development decision-making. Washington, D.C.: World Bank Group.

Aghion

Ljungqvist

Howitt

, et al. (1998) Endogenous Growth Theory. Cambridge, MA: MIT Press.

Amsden

(1994) East-Asian Miracle: Economic growth and public policy. World Development 22(4): 614–670.

Capoccia

(2015) Critical junctures and institutional change. In: Mahoney

Thelen

(eds) Advances in Comparative-historical Analysis. Cambridge: Cambridge University Press, pp. 147–149.

Capoccia

Kelemen

(2007) The study of critical junctures: Theory, narrative, and counterfactuals in historical institutionalism. World Politics 59(3): 341–369.

Carrubba

Gabel

Helmke

, et al. (2015) When Parchment Barriers Matter: De Jure Judicial Independence and the Concentration of Power. manuscript (on file with authors).

Cherry

(2005) ‘Big Deal’ or big disappointment? The continuing evolution of the South Korean developmental state. The Pacific Review 18(3): 327–354.

Choi

Kang

(2014) Competition law meets corporate governance: Ownership structure, voting leverage, and investor protection of large family corporate groups in Korea. Peking University Transnational Law Review 2(2): 411–444.

10.

Dixit

(2009) Governance institutions and economic activity. The American Economic Review 99(1): 3–24.

11.

Evans

Rauch

(1999) Bureaucracy and growth: A cross-national analysis of the effects of “Weberian” state structures on economic growth. American Sociological Review 64(5): 748–765.

12.

Feenstra

Hamilton

Lim

(2002) Chaebol and catastrophe: A new view of the Korean business groups and their role in the financial crisis. Asian Economic Papers 1(2): 1–45.

13.

Freedman

(2005) Economic crises and political change: Indonesia, South Korea, and Malaysia. Asian Affairs: An American Review 31(4): 232–249.

14.

Glaeser

La Porta

Lopez-de-Silanes

, et al. (2004) Do institutions cause growth? Journal of Economic Growth 9(3): 271–303.

15.

Haggard

(2004) Institutions and growth in East Asia. Studies in Comparative International Development 38(4): 53–81.

16.

Haggard

MacIntyre

Tiede

(2008) The rule of law and economic development. Annual Review of Political Science 11: 205–234.

17.

Hans-Joachim

(2015) Formal and informal institutions. In: Gandhi

Ruiz-Rufino

(eds) Routledge Handbook of Comparative Political Institutions. London: Routledge, pp. 56–69.

18.

Hayo

Voigt

(2007) Explaining de facto judicial independence. International Review of Law and Economics 27(3): 269–290.

19.

Helmke

Levitsky

(2004) Informal institutions and comparative politics: A research agenda. Perspectives on Politics 2(4): 725–740.

20.

Jung

J-Y

(2011) How a powerful bureaucracy fell: The abolition of the Economic Planning Board in South Korea. Korea Observer 42(4): 551–581.

21.

Kihl

(2005) The Past as Prologue: President Kim Dae Jung’s Legacy and President Roh Moo Hyun’s Policy Issues and Future Challenges. In: Mansourov

(ed) ROK Turning Point. Honolulu: Asia-Pacific Center for Security Studies, pp. 158–183.

22.

Kohli

(2004) State-directed Development: Political Power and Industrialization in the Global Periphery. New York: Cambridge University Press.

23.

Krueger

(1995) East Asian experience and endogenous growth theory. In Ito

Krueger

(eds) Growth Theories in Light of the East Asian Experience, NBER-EASE Volume 4. Chicago: University of Chicago Press, pp. 9–36.

24.

Krugman

(1998) What happened to Asia? Unpublished paper. Massachusetts Institute of Technology (January).

25.

Kurtz

Schrank

(2007) Growth and governance: Models, measures, and mechanisms. Journal of Politics 69(2): 538–554.

26.

Lee

P-S

(2000) Economic crisis and chaebol reform in Korea. School of Business Administration, Korea University Discussion Paper no. 14. Oct. 2000.

27.

Lee

(2008) The politics of chaebol reform in Korea: Social cleavage and new financial rules. Journal of Contemporary Asia 38(3): 439–452.

28.

Leipziger

(2001) Lessons from East Asia. Ann Arbor: University of Michigan Press.

29.

Levchenko

(2007) Institutional quality and international trade. The Review of Economic Studies 74(3): 791–819.

30.

(2003) Relation-based versus rule-based governance: An explanation of the East Asian Miracle and Asian Crisis. Review of International Economics 11(4): 651–673.

31.

(2013) China’s (painful) transition from relation-based to rule-based governance: When and how, not if and why. Corporate Governance: An International Review 21(6): 567–576.

32.

(2014) The inevitable and difficult transition from relation-based to rule-based governance in China. Modern China Studies 21(1): 145–171.

33.

Linzer

Staton

(2015) A global measure of judicial independence, 1948–2012. Journal of Law and Courts 3(2): 223–256.

34.

North

(1990) Institutions, Institutional Change and Economic Performance. New York: Cambridge University Press.

35.

North

(1991) Institutions STÖR. The Journal of Economic Perspectives 5(1): 97–112.

36.

Okabe

(2015) Reacting to financial crises. In: Pempel

Tsunekawa

(eds) Two Crises, Different Outcomes: East Asia and Global Finance. Ithaca, NY: Cornell University Press, pp. 90–109.

37.

Park

(2011) Revisiting the South Korean developmental state after the 1997 financial crisis. Australian Journal of International Affairs 65(5): 590–606.

38.

Pempel

Tsunekawa

(2015) Two crises, two outcomes. Two Crises, Different Outcomes: East Asia and Global Finance. Ithaca, NY: Cornell University Press, pp. 17–38.

39.

Rajan

Zingales

(1998) Which capitalism? Lessons from the East Asian crisis. Journal of Applied Corporate Finance 11(3): 40–48.

40.

Rodrik

Subramanian

Trebbi

(2004) Institutions rule: The primacy of institutions over geography and integration in economic development. Journal of Economic Growth 9(2): 131–165.

41.

Soifer

(2012) The causal logic of critical junctures. Comparative Political Studies 45(12): 1572–1597.

42.

Soifer

(2013) State power and the economic origins of democracy. Studies in Comparative International Development 48(1): 1–22.

43.

Steinmo

Thelen

(1992) Structuring Politics: Historical Institutionalism in Comparative Analysis. New York: Cambridge University Press.

44.

Stiglitz

(1996) Some lessons from the East Asian miracle. The World Bank Research Observer 11(2): 151–177.

45.

Torgler

Schneider

(2007) Shadow economy, tax morale, governance and institutional quality: A panel analysis. School of Economics and Finance Discussion Papers and Working Papers Series 210, School of Economics and Finance, Queensland University of Technology.

46.

Wade

(1994) Resolving the state-market dilemma in East Asia. Advanced Development Management Program, Institute of Comparative Culture, Sophia University.

47.

Weingast

(1995) The economic role of political institutions: Market-preserving federalism and economic development. Journal of Law, Economics, & Organization 11(1): 1–31.

48.

Weyland

(2004) The Politics of Market Reform in Fragile Democracies: Argentina, Brazil, Peru, and Venezuela. Princeton, NJ: Princeton University Press.

49.

White

Wade

(1988) Developmental states and markets in East Asia: An introduction. In: White

(ed) Developmental States in East Asia. Basingstoke: Macmillan Press, pp. 1–29.

50.

Williamson

(2009) Informal institutions rule: Institutional arrangements and economic performance. Public Choice 139(3–4): 371–387.

51.

Williamson

Kerekes

(2011) Securing private property: Formal versus informal institutions. The Journal of Law and Economics 54(3): 537–572.

52.

Woo

J-E

Woo-Cumings

(1991) Race to the Swift: State and Finance in Korean Industrialization. New York: Columbia University Press.

53.

Yanagimachi

(2004) Chaebol reform and corporate governance in Korea. Working Paper. Keio University.

Towards rule-based institutions and economic growth in Asia? Evidence from the Asian Financial Crisis 1997–1998

Abstract

Keywords

Introduction

Institutions, the Asian Financial Crisis, and IMF reform

Different mechanisms of economic growth in Asia and the advanced West

Causes of the Asian Financial Crisis

Conditionality for the IMF’s bailout reforms

Institutional change, critical juncture, and the Asian Financial Crisis

Empirical strategy

Empirical model

Model specification

Data

Results

Main findings

Robustness check

Exploring the mechanism for lack of reform

Conclusion

Footnotes

Author’s note

Declaration of conflicting interests

Funding

ORCID iD

Note

References