Elsevier

Journal of Monetary Economics

Volume 85, January 2017, Pages 69-86
Journal of Monetary Economics

After the tide: Commodity currencies and global trade

https://doi.org/10.1016/j.jmoneco.2016.11.005Get rights and content

Abstract

The decade prior to the Great Recession saw a boom in global trade and rising transportation costs. High-yielding commodity exporters׳ currencies appreciated, boosting carry trade profits. The Global Recession sharply reversed these trends. We interpret these facts with a two-country general equilibrium model that features specialization in production and endogenous fluctuations in trade costs. Slow adjustment in the shipping sector generates boom–bust cycles in freight rates and, as a consequence, in currency risk premia. We validate these predictions using global shipping data. Our calibrated model explains about 57% of the narrowing of interest rate differentials post-crisis.

Introduction

The decade prior to the Great Recession saw a boom in global trade, including a rapid rise in commodity prices, trade volumes, and, consequently, in the cost of transporting goods around the world. At the same time currencies of commodity-exporting currencies appreciated, boosting the carry trade profits in foreign exchange markets (commodity currencies typically earn higher interest rates, making them attractive to investors).1 The onset of the Global Recession led to a sharp reversal in all of these trends, with only a weak recovery subsequently. We interpret these facts through the lens of an international asset pricing model, focusing on two groups of countries whose currencies represent the two sides of a typical carry trade strategy. The first group consists of developed countries that are major exporters of basic commodities (Australia, Canada, New Zealand, and Norway) – the typical “investment” currencies.2 The second group consists of developed economies that primarily export complex manufactured goods (the Euro zone, Japan, Sweden, and Switzerland), typically seen as providing “funding” currencies for exchange rate speculation due to their historically low interest rates.

Building on Ready et al. (2016), we develop a two-country general equilibrium model that features complete financial markets and specialization in trade. The key friction that we emphasize in this paper is the slow adjustment of capacity in the shipping sector, which results in highly variable costs of international trade (or at least their component that is attributed to shipping). The model can jointly account for the dynamic behavior of real exchange rates and interest rates (and therefore carry trade returns), commodity prices, and shipping costs. All of these series exhibit a sharp drop during the crisis, followed by very slow recovery, with shipping costs being the most sluggish. Our quantitative model implies that expected returns on the commodity currency carry trade fall by about one half in the wake of the Great Recession.

The model is designed to capture the carry trade in currency markets. In the model, differences in average interest rates and risk exposures between countries that are net importers of basic commodities (“producer countries”) and commodity-exporting countries (“commodity countries”) are rationalized by appealing to a natural economic mechanism: trade costs.3 We model trade costs by considering a simple model of time-varying trade frictions. At any time the cost of transporting a unit of good from one country to the other depends on the aggregate shipping capacity available. While the capacity of the shipping sector adjusts over time to match the demand for transporting goods between countries, it does so slowly, e.g., due to gestation lags in the shipbuilding industry. In order to capture this intuition we assume that the marginal cost of shipping an extra unit of good is increasing – i.e., trade costs in our model are convex, as in Ready et al. (2016).

Convex shipping costs imply that the sensitivity of the commodity country to world productivity shocks is lower than that of the country that specializes in producing the final consumption good, simply because it is costlier to deliver an extra unit of the consumption good to the commodity country in good times, but cheaper in bad times. Therefore, under complete financial markets, the commodity country׳s consumption is smoother than it would be in the absence of trade frictions, and, conversely, the producer country׳s consumption is riskier. Since the commodity country faces less consumption risk, it has a lower precautionary saving demand and, consequently, a higher interest rate on average, compared to the country producing manufactured goods. Since the commodity currency is risky – it depreciates in bad times from the perspective of the producer country׳s consumer – it commands a risk premium. Therefore, the interest rate differential is not offset on average by exchange rate movements, giving rise to a carry trade. The role of slowly adjusting shipping capacity, which we emphasize here, is to amplify these fluctuations in trade costs, and, therefore, in exchange rates, especially during cyclical transitions. When global output expands due to rising productivity, trade costs increase sharply until shipping capacity catches up with output (and exports); a contraction that follows such a build-up produces a sharp drop in transport costs as accumulated shipping capacity is large relative to the amount of goods being shipped.

In order to evaluate the model׳s ability to generate quantitatively reasonable magnitudes of currency risk premia and interest rates we calibrate it by allowing for the possibility of very large jumps in productivity – i.e., rare disasters, as in the literature on the equity premium puzzle (e.g., Longstaff and Piazzesi, 2004, Barro, 2006, Gabaix, 2012, Wachter, 2013). The calibrated model is able to account for the observed interest rate differentials and average returns on the commodity currency carry trade strategies without overstating consumption growth volatility, even in samples that contain disasters, or implying an unreasonably high probability of a major disaster.4

We use our model as a laboratory for understanding the behavior of commodity currencies around the Great Recession. We feed in a series of productivity shocks observed in the data. Over the period 2002–2006 these shocks generate a boom in commodity prices (as commodity supply struggles to catch up quickly) and a rise in global shipping costs (as shipping capacity also lags behind). The commodity country exchange rate appreciates as well, yielding high carry trade profits, essentially matching those observed in the data. The latter result is not mechanical, as in the presence of complete financial markets terms of trade do not drive exchange rates; rather, this is due to the fact that increasing trade costs make markets more segmented, with marginal utility of producer country (“Japan”) consumers fall faster than that of commodity country (“Australia”). The ensuing global crisis is represented in our model by a large negative productivity shock in the producer country (we abstract from demand shocks or financial frictions for simplicity). As a result, output and trade in the final good collapse, as do commodity prices and shipping costs. The commodity currency depreciates due to a sharp spike in the marginal utility in “Japan” relative to that of “Australia”, generating large losses for the currency carry trade. Since shipping capacity is very slow to adjust, trade costs remain depressed even as output and trade recover. Interest rate differentials and expected carry trade returns also decrease as low trade costs imply a greater degree of risk sharing (i.e., a closer alignment of marginal utilities, and consequently less scope for a currency risk premium). We show that all of these predictions are consistent with the empirical evidence for the countries that we consider.

Our analysis sheds a new light on the role of time-varying transport costs in international trade. In our model trade costs increase in (global) good times endogenously, since that is when exports tend to rise. This is consistent with arguments in Hummels (2007) and papers cited therein, emphasizing the effects of port congestion and delays in shipping and the role of fuel costs, which in the recent decades have behaved procyclically, as well as evidence on the value of speed of shipment analyzed in Hummels and Schaur (2013). It is also corroborated by the empirically observed behavior of the shipping cost indices that we consider as the most explicit (albeit narrow) measures of transportation costs. Other models in international finance instead assume that trade costs increase (exogenously) in bad times, potentially due to a tightening of trade credit – e.g., Maggiori (2012). We do not need to take a stand on the role of trade finance in the trade collapse during the Great Recession.5 While our model of international trade is relatively stylized compared to the recent models that focus on understanding trade frictions at the micro-level (e.g., Arkolakis, 2010, Alessandria et al., 2013), our contribution is to highlight the importance of shipping capital for the dynamics of global trade, international risk sharing, and real exchange rates.

Section snippets

Motivating evidence

Our approach builds on the evidence in Ready et al. (2016) that countries whose primary exports are basic commodities exhibit qualitatively different behavior of macroeconomic aggregates than countries that concentrate in exporting complex manufactured goods. As argued by Lustig et al. (2011), persistent differences in countries׳ exposures to global shocks drive the bulk of currency risk premia earned in foreign exchange markets. These differences lead to persistent interest rate differentials

Model

There are two countries each populated by a representative consumer endowed with CRRA preferences over the same consumption good with identical coefficients of relative risk aversion γ and rates of time preference ρ. The two countries specialize in the production of a single good and trade with each other through interactions with a shipping industry and commodity traders.

The “commodity” country produces a basic input good using a linear production technologymaxlcP*zclcwclc.One unit of

Quantitative analysis

So far we have only explored the qualitative implications of our model. We now turn to quantitative analysis. Ideally, we would like to calibrate the model parameters to closely match a set of key empirical moments. The fact that the model features only two countries (each completely specialized in producing one kind of good) makes such a moment-matching exercise challenging. In order to circumvent this challenge we treat the key manufacturing-good exporting countries as a group, assuming that

Great Recession: model and data

In order to gauge the severity of the Great Recession׳s impact on shipping capacity and globalization post-recession, we perform a simulation exercise where we take our benchmark calibration as the structure of our economy and use empirical data on realized growth rates of productivity and shipping capital to proxy for the fundamental shocks.

Specifically, we take industrial production growth rates for our portfolio of producer countries and commodity countries as proxies for Δzc and Δzp.14

Conclusion

The Great Recession generated large movements in exchange rates and currency carry trade returns. We use the calibrated model to account for the behavior of the relevant economic time series before and after the Great Recession. The run-up in global productivity prior to the recession leads to increases in commodity prices (as commodity supply struggles to catch up quickly) and in global shipping costs (as shipping capacity also lags behind). A widening wedge between the marginal utilities in

Acknowledgments

We benefitted from comments and suggestions by Andy Abel, George Alessandria (the editor), Mathieu Taschereau-Dumouchel, Doireann Fitzgerald (the discussant), Ivan Shaliastovich, and conference participants at the Carnegie-Rochester-NYU Conference on Public Policy. Roussanov thanks Wharton Global Initiatives for research support.

References (47)

  • M. Amiti et al.

    Exports and financial shocks

    QJE

    (2011)
  • J.E. Anderson et al.

    Trade costs

    J. Econ. Lit.

    (2004)
  • C. Arkolakis

    Market penetration costs and the new consumers margin in international trade

    J. Polit. Econ.

    (2010)
  • D. Backus et al.

    International real business cycles

    J. Polit. Econ.

    (1992)
  • D. Backus et al.

    Disasters implied by equity index options

    J. Financ.

    (2011)
  • G.S. Bakshi et al.

    Equilibrium valuation of foreign exchange claims

    J. Financ.

    (1997)
  • R.J. Barro et al.

    Macroeconomic crises since 1870

    Brook. Pap. Econ. Act.

    (2008)
  • R. Barro

    Rare disasters and asset markets in the twentieth century

    Q. J. Econ.

    (2006)
  • H. Bessembinder et al.

    Mean reversion in equilibrium asset pricesevidence from the futures term structure

    J. Financ.

    (1995)
  • Burnside, C., Eichenbaum, M., Kleshchelski, I., Rebelo, S., 2008. Can Peso Problems Explain the Returns to the Carry...
  • Y.-c. Chen et al.

    Can exchange rates forecast commodity prices?

    Q. J. Econ.

    (2010)
  • Chernov, M., Graveline, J., Zviadadze, I., 2012. Crash Risk in Currency Returns. Working...
  • Colacito, R., Croce, M.M., Ho, S., Howard, P., 2013. BKK the EZ Way: An International Production Economy with Recursive...
  • Cited by (26)

    • Commodity price effects on currencies

      2023, Journal of International Money and Finance
      Citation Excerpt :

      These modifications, coupled with altered economic conditions, have triggered changes in the global financial market in general and the global commodity market in particular. For instance, Adams and Glück (2015), Ordu, et al. (2018), and Ready, et al. (2017a) noted different behaviors of commodity price indexes before and after the GFC. Commodity financialization gained momentum in the post-GFC era, thereby making commodities a viable asset class for institutional and individual investors (Adams and Glück, 2015; Baker, 2021; Basak and Pavlova, 2016; Henderson, et al., 2015; Ordu, et al., 2018).

    • Origins of international factor structures

      2023, Journal of Financial Economics
    • The time-varying risk price of currency portfolios

      2022, Journal of International Money and Finance
    • Entangled risks in incomplete FX markets

      2021, Journal of Financial Economics
      Citation Excerpt :

      Then, the exchange rate can be disconnected from consumption while it is still equal to the ratio of country-specific SDFs. Examples are models of long-run risk (Colacito and Croce, 2011; Bansal and Shaliastovich, 2012; Colacito and Croce, 2013; Colacito et al., 2018a; 2018b), habit formation preferences (Verdelhan, 2010; Stathopoulos, 2017), disaster risks (Maggiori and Gabaix, 2015; Farhi and Gabaix, 2016; Farhi et al., 2015), frictions in shipping (Ready et al., 2017b; 2017a), trade network centrality (Richmond, 2019; Jiang and Richmond, 2020), government fiscal conditions (Jiang, 2021), heterogeneous agents (Kollmann, 2009; Fang, 2018), segmented asset markets (Chien et al., 2020), and financial intermediaries (Fang and Liu, 2020). The second approach uses market incompleteness to introduce a wedge between the exchange rate and the ratio of SDFs (Zapatero, 1995; Sarkissian, 2003; Dumas et al., 2003; Chaieb and Errunza, 2007; Pavlova and Rigobon, 2007; Favilukis et al., 2015; Bakshi et al., 2018; Lustig and Verdelhan, 2019; Sandulescu et al., 2020).

    • Curve momentum in currency markets

      2021, Finance Research Letters
      Citation Excerpt :

      The findings imply that curve momentum strategies yield positive excess return and spot premium at both country-level and portfolio-level. Interestingly, the curve momentum return crashes the during subprime crisis, corresponding to the theoretical results of Brunnermeier and Pedersen (2009), which also reveals that the structure of FX market changed after 2008 (Du et al., 2018; Ready et al., 2017). Second, we find that the cross-curve momentum signal is a powerful predictor for both excess return and spot premium.

    View all citing articles on Scopus
    View full text