The Political Economy of Exchange Rates and Export-Led Industrialization in South Korea*

ABSTRACT

South Korea’s economic trajectory has been one of the most successful stories of post-war catch-up industrialization. The Korean developmental state has used various tools, including the exchange rate, to become an export powerhouse in the twenty-first century. However, the export-led growth strategy has had a distributive conflict with Korea’s working class. Based on post-Keynesian and structuralist economics, this paper assesses the political economy of exchange rate depreciation and export-led industrialization in South Korea. I show that exports as a share of GDP have constantly risen to become the most important growth engine in the Korean economy. This long-term trend, however, has coincided with a decline in the labor income share in South Korea. I formulate the structuralist macroeconomic model to explain dynamics of real wages, real exchange rates, exports, and output. I finally use a vector autoregression (VAR) analysis to show that exchange rate depreciations during modern South Korea (1989Q1–2023Q1) are inflationary and contractionary in the short run. Currency depreciations have ambiguous effects on exports, but they clearly reduce output gaps. On the other hand, an increase in nominal wages is expansionary or varies positively with output gaps.

KEYWORDS:

• Time series analysis
• South Korea
• exchange rates
• income distribution
• contractionary devaluation

JEL CODES:

• C32
• E12
• F41
• O11
• O53

Acknowledgements

I want to use this opportunity to acknowledge and thank the editor-in-chief of the Review of Political Economy, Louis-Philippe Rochon, and two anonymous referees for their valuable comments and suggestions.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Notes

1 The slope of the RER Response curve equals $\left[\left(1-\alpha \right)b\right]/\left[a\left(\alpha -1\right)-\text{β}\right]$. The slope of the Real Wage Response curve equals $-b/a$. As long as $\text{β}>0$, which is already given as $0<a,b,\alpha ,\text{β},\gamma <1$, the slope of the Real Wage Response curve is steeper than the slope of the RER Response curve.

2 Given that $\mathrm{\partial }\dot{\omega }/\mathrm{\partial \omega }=\left(\gamma -1\right)b$, $\mathrm{\partial }\dot{\omega }/\mathrm{\partial \rho }=\left(\gamma -1\right)a$, $\mathrm{\partial }\dot{\rho }/\mathrm{\partial \rho }=\left(\alpha -1\right)b$, and $\mathrm{\partial }\dot{\rho }/\mathrm{\partial \omega }=\left(\alpha -1\right)a-\text{β}$.

3 The implications of injections/leakages, elasticities, and Marshall-Lerner conditions are further elaborated in Taylor (2004, pp. 254–257).

4 The J-curve implies the movement of a trade balance over a short period of time after a devaluation. According to Krueger (1983), because goods are in transit and the contracts are already made, the price elasticity does not function during the short period of time, which makes a trade balance fall at first. Over time, once the price elasticity kicks in, the trade balance will begin to improve. Bahmani-Oskooee (1985) finds that, based on the quarterly data of the period 1973–80, it takes 3 quarters after a devaluation for Korea to improve the trade balance.

5 The rate of employment is much more common for this type of model. However, I follow Lance Taylor’s formation (1991, 2004, 2005) in this study. In those works, capacity utilization is used instead of the rate of employment. Capacity utilization can be an alternative variable to measure whether the economy is inflationary or not.

6 This scenario is likely at least in the short to medium run until export capacity responds through the price incentives in the longer run.

7 In contrast to some previous literature, fiscal and monetary policy are excluded from the VAR estimation. Variables such as fiscal deficit, public debt to GDP, or interest rate measures are therefore ignored from this paper. I thank the reviewer for pointing out this importance.

8 This also has a theoretical implication that, as the export response to the exchange rate is weak, the dynamics in the model Figure 5 are unstable.