Foreign currency borrowing, exports and firm performance: evidence from a currency crisis

ABSTRACT

This paper develops a simple signaling model of foreign currency borrowing that yields predictions about firm survival and performance during a currency crisis. Using a large panel of firm level data for South Korea we offer empirical support for many of the predictions of our model, while others support predictions that cannot be tested using our data. Our paper demonstrates that although firms that borrow in foreign currency are more likely to exit after the currency collapses, those that continue to produce perform better. Among them, the best performers are exporters whose foreign sales are more competitively priced under a devalued currency.

KEYWORDS:

• Currency crisis
• exports
• foreign currency borrowing

JEL CLASSIFICATIONS:

• F34
• F41
• G21
• L25

Acknowledgements

We thank Mike Bleaney, John Driffill, Hans Genberg, Alessandra Guariglia, Junhan Kim, Tae-Hwan Kim, Mihye Lee and seminar participants at the Hong Kong Institute of Monetary Research, the University of Nottingham at Ningbo, China, the Research Workshop on Finance and Development at Limassol, Cyprus and the European Trade Study Group, Munich for comments on this paper. Mizen acknowledges with thanks the financial support of the Hong Kong Institute for Monetary Research. The views expressed herein are those of the authors and do not necessarily reflect the official views of the Bank of Korea or the Central Bank of the Republic of Turkey.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Simona Mateut http://orcid.org/0000-0002-3737-5952

Paul Mizen http://orcid.org/0000-0002-6193-0872

Notes

1. In many respects South Korea is an apt comparison for the recent global financial crisis. The East-Asian crisis brought about a 6.7% contraction of GDP growth in 1998, and a 40% reduction in fixed investment – the sharpest decline in real activity since 1950 – which is comparable in many respects to the severity of the recent global financial crisis. After short-term rates fell dramatically with the devaluation of the South Korean won, credit to the private sector declined, and banks were subject to greater, externally imposed regulation, further diminishing the incentives to lend. The devaluation in the currency provided a competitive advantage to exporters, however, as this paper documents. See Sohn (2010) for an overview of the reforms implemented on the financial system after the crisis.

2. The broad literature on signaling in financial markets is comprehensively reviewed in Tirole (2006, Ch. 6).

3. Data stop at 2006 to ensure that our analysis is not affected by the global financial crisis that began in Fall 2007.

4. All of these papers are purely theoretical.

5. Given that the sample we use to test our model includes unlisted firms whose owners might not have the same diversification opportunities as those of listed firms, the assumption of risk neutrality is not necessarily without consequence. When we discuss our main results we also consider the implications of allowing for risk aversion.

6. We can allow for different values across firm types as long as the difference is not too large so that type firms have an incentive to strategically default.

7. We will demonstrate that for the existence of a signaling equilibrium we do not need to impose any further restrictions on revenues.

8. Strictly speaking in our simple model firms fail because of negative idiosyncratic shocks and therefore the rates of failure for the two types of firms are not affected by the depreciation of the currency. But it is possible to consider a more general environment allowing for more states in which high-productivity firms can survive in more states than low-productivity firms when both types of firms borrow in foreign currency. Put differently, the cost of default is higher for low-productivity firms to allow separation of the two types to be feasible.

9. The sample in Kim, Tesar, and Zhang (2015) includes firms from different economic sectors observed over the period 1994–1999. Using additional information, unavailable to us, regarding firms that submitted a notification of closing business to the court system, they are able to identify liquidated and surviving firms.

10. To control for firm fixed effects we estimate the model in first-differences, which is equivalent to taking deviations from the mean (the fixed effects estimator) when there are only two time periods.

11. Firms above the nominal asset threshold of 7 bn won are required to report annual financial statements; firms below the threshold may voluntarily report if they wish to.

12. For example, Kim, Tesar, and Zhang (2015) find that the net worth of firms that carried a lot of short-term foreign currency debt prior to the crisis dropped significantly during the crisis.

13. There are other issues to consider in this relationship. For example, to address the potential endogeneity of our export measures we use their lagged values instead and find that these results perfectly resemble those reported in Table 5. Some authors have also paid attention to the causal relationship between foreign currency loans and exporting (e.g. Minetti and Zhu 2011), who find access to finance is a significant determinant of a firm's decision to export. Our model is silent about the direction of causality but our empirical results indicate that exporters are more likely to access foreign currency loans than domestic firms.

14. A large international trade literature, following the seminal work by Melitz (2003), makes a positive link between entry to export markets and firm size through sunk costs (Bernard et al. 2003; Bernard and Jensen 2004; Campa 2004; Helpman, Melitz, and Yeaple 2004; Roberts and Tybout 1997; Roberts, Sullivan, and Tybout 1997, 2003). Empirical support for this view is cited in Girma, Greenaway, and Kneller (2004) and Greenaway, Guariglia, and Kneller (2007) for firms from Germany, Italy, Latin America, Spain, the UK and the US. Aw and Hwang (1995) and Aw, Chung, and Roberts (2000) draw the same conclusions from a sample of Taiwanese and South Korean firms.

15. Bernini, Guillou, and Bellone (2015) explore the relationship between export quality and leverage.

16. We thank a referee for suggesting this exercise. Kim, Tesar, and Zhang (2015) use a similar specification and opt for estimating two cross-sectional equations, one for the crisis and one for the pre-crisis period. We exploit the large panel dimension of our data and estimate with firm fixed effects.

17. See https://www.jbic.go.jp/en/efforts/result-abmi.

Additional information

Funding

Mizen acknowledges with thanks the financial support of the Hong Kong Institute for Monetary Research.

Notes on contributors

Spiros Bougheas

Spiros Bougheas is Professor of Economics, School of Economics, University of Nottingham.

Hosung Lim

Hosung Lim is Senior Economist at the Bank of Korea.

Simona Mateut

Simona Mateut is Associate Professor in Financial Economics at Nottingham University Business School.

Paul Mizen

Paul Mizen is Professor of Monetary Economics & Director, Centre for Finance, Credit and Macroeconomics; he is also the Chairman, Money, Macro and Finance Research Group.

Cihan Yalcin

Cihan Yalcin is Senior Economist, Structural Economic Research Department, The Central Bank of Turkey.