This is a summary of "Interconnectedness and Systemic Risk: A Comparative Study Based on Systemically Important Regions," by Lei Fang, Jiang Cheng, and Fang Su, published in the Pacific Basin Finance Journal.
The authors investigate the association between interconnectedness and systemic risk and find that a multi-centered structure, rather than a single pivotal center, may reduce global systemic risk. Although interconnectedness is positively related to systemic risk, a higher GDP, larger population, and stronger currency may help a country to evolve into a systemically important region (SIR) that is immune to systemic risk.
What Is the Investment Issue?
During the financial crisis of 2007–2009, the risks of subprime mortgage loans spread rapidly to markets around the world through the financial system. Similar contagions have been observed in other integrated regions, such as the Asian Financial Crisis in 1997. Although closer economic cooperation improves efficiency, opening up the economy also makes it more vulnerable to external risks. Interconnectedness is recognized as an important determinant of systemic risk.
In this era of economic interconnectedness, is higher systemic risk unavoidable? The authors argue that a regional-level increase in systemic risk could reduce aggregate global-level systemic risk, because the risks from interconnectedness are less concentrated. If this is true, a multi-centered structure of interconnection could help reduce global aggregate risk.
How Did the Authors Conduct This Research?
The authors seek the strongest and most direct network connections among regions. Assuming each region is a node, the authors use the minimum spanning tree (MST) method to identify each node in the network. Systemic risk is believed to be transmitted through these nodes in the network. Foreign direct investment (FDI) is used to proxy interconnectedness because it reflects multiple aspects of the economy, such as economic growth and exports. FDI data are from the Economist Intelligence Unit (EIU) Country Data reports, which covers 60 regions from 1989 to 2015.
Using the MST method, the authors derive a global financial network. If a region is more interconnected with others, it will have more “lines” connecting it to others in the network, showing that the region is a pivot in the network. The authors define the regions with the top 10% most connected lines as SIRs; the other regions are defined as non-SIRs (NSIRs). The number of SIRs for each year between 1989 and 2015 is identified. The authors then use the Herfindahl–Hirschman Index (HHI) to measure the degree of concentration of the interconnectedness.
Next, the impact of interconnectedness on systemic risk is measured. To proxy systemic risk, the authors use the EIU Country Risk Score. This score comprehensively reflects sovereign risk, currency risk, and banking sector risk. By using the orthogonal pulse function, the direction and time lag of the effect of interconnectedness on systemic risk at a global level and regional level are analyzed. Finally, the authors decompose the difference between SIRs and NSIRs to make recommendations for NSIRs hoping to evolve into SIRs.
What Are the Findings and Implications for Investors and Investment Professionals?
The authors find that the global financial network has developed from a limited number of large centers in the 1990s to numerous smaller centers in the 2010s. In 1989, the United States was the only “super center,” with 24 connected lines; 2015 saw more centers, including the Netherlands, France, Japan, Germany, and China. These centers are smaller, with fewer connected lines than the United States—ranging from 4 to 11 in 2015. The HHI also decreases over time, which generally confirms that the interconnectedness is becoming less concentrated, consistent with the observation that the emergence of multiple SIRs reduces the concentration.
The analysis also confirms that a higher HHI produces a significant positive effect on global systemic risk. The risk diminishes to insignificance after five years. This result supports the hypothesis that the more concentrated the network, the higher the global systemic risk.
At a regional level, however, the effects of interconnectedness on systemic risk are significant only on NSIRs. Increasing interconnectedness does not increase the systemic risk of SIRs. The authors believe that SIRs have more channels for transferring risk, which suggests that the positive effect of connectedness cancels out the negative effect.
To reduce its systemic risk, an NSIR can either de-integrate from external economies or evolve itself into an SIR. The latter is obviously a better choice. The authors find that SIRs have relatively higher GDP, larger population, and stronger currencies than NSIRs. It may be beneficial for NSIRs to develop their strengths in these three areas.
The contagion effect has become one of the major concerns of economic and financial integration. Investors should consider those countries that can better defend against global crisis, particularly those with a large foreign reserve, strict credit control, and strong cooperation with neighbors.