The impact of yuan internationalization on the stability of the international monetary system

ARTICLE IN PRESS Journal of International Money and Finance ■■ (2015) ■■–■■

Contents lists available at ScienceDirect

Journal of International Money and Finance j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / j i m f

The impact of yuan internationalization on the stability of the international monetary system Agnès Bénassy-Quéré a,*, Yeganeh Forouheshfar b a b

Paris School of Economics, University Paris 1, France University Paris-Dauphine, France

A R T I C L E

I N F O

Article history: Available online JEL Classification: F31 F33

Keywords: China Yuan Exchange-rate regime Euro Dollar

A B S T R A C T

We study the implication of yuan internationalization on the stability of the international monetary system. More specifically, we use a three-country, three-currency portfolio model to analyze the impact of yuan internationalization on exchange rates in the event of trade shocks, with stock-flow adjustment of the net foreign asset positions. We show that the internationalization of the yuan would lessen the response of floating exchange rates to asymmetric trade shocks as well as attenuate the distortionary impact of China keeping its currency pegged to the dollar. Conversely, yuan internationalization would amplify the impact of trade shocks on net foreign asset positions, albeit to a limited extent. © 2015 Elsevier Ltd. All rights reserved.

1. Introduction The US dollar has been the key currency of the international monetary system (IMS, hereafter) since World War II. The yen never succeeded in challenging the US currency as an international means-ofpayment, unit-of-account or store-of-value currency. As for the euro, it has emerged mostly as a diversification and as a regional currency (European Central Bank, 2014). Such resilience on the part of the dollar was benign as long as the US economy was clearly dominant in terms of GDP, trade and financial markets: monetary pegs and reserve accumulation did not weigh much on global trade and capital flows. However the share of the United States in the global economy fell from 39% in 1960 to 22% in 2014. In 2014, countries running a fixed or managed exchange-rate regime (de jure or de facto)

* Corresponding author. Tel.: +144078219. E-mail address: [email protected] (A. Bénassy-Quéré). http://dx.doi.org/10.1016/j.jimonfin.2015.05.004 0261-5606/© 2015 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Agnès Bénassy-Quéré, Yeganeh Forouheshfar, The impact of yuan internationalization on the stability of the international monetary system, Journal of International Money and Finance (2015), doi: 10.1016/ j.jimonfin.2015.05.004

ARTICLE IN PRESS 2

A. Bénassy-Quéré, Y. Forouheshfar / Journal of International Money and Finance ■■ (2015) ■■–■■

with respect to the US dollar accounted for 18% of world GDP, with China alone (which the IMF classified as in a de facto crawling peg arrangement with the dollar) accounting for 12% of world GDP. 1 The mismatch between a unipolar IMS and a multipolar real economy has at times been singled out as a key ingredient of the macroeconomic environment that led to the 2008–09 financial crisis (see e.g. Ivashina et al., 2012). Leading policy-makers have also argued that a unipolar IMS creates a deflationary bias, with all countries other than the key-currency issuer wishing to accumulate foreignexchange reserves through current-account surpluses in order to self-insure against a reversal of gross capital inflows (see United Nations, 2009). However, the policy debate is mainly concerned with the Triffin dilemma (Triffin, 1960), which was initially phrased to show the internal inconsistency of the Bretton Woods system. It has since been revisited by Zhou (2009) and Fahri et al. (2011) in the context of the post-Bretton Woods system and taking into account emerging economies. According to Farhi et al., the growth of emerging economies and their appetite for liquid, riskless assets increase the demand for international liquidity, putting downward pressure on US interest rates and on the US current account. At some stage, international investors will either lose confidence in US solvency or fear massive monetization of US bonds, which would trigger a crash of the dollar. To avoid such an outcome, it would be advisable to develop alternative sources of international liquidity, either through the internationalization of other currencies or the development of Special Drawing Rights (Mateos y Lago et al., 2009). Another strand of the literature focuses on the incentive a hegemonic country may have to ensure the stability of the system. Kindleberger (1973, 1981) argues that this could explain why hegemonic systems could be more stable, whereas Eichengreen (1987) suggests instead there is a risk the hegemon will abuse its position. According to Cohen (2009), monetary power fragmentation could involve both economic risks (e.g. increasingly antagonistic relationships between currency blocs, possibly leading to de-globalization) and geopolitical ones (e.g. weaker support for the US dollar coming along with weaker military protection from the United States). Since the mid-1990s, China has taken significant steps toward the progressive internationalization of the yuan: it has allowed domestic exporters to invoice their cross-border sales in yuan; it has cautiously opened the gate to foreign capital inflows; it has developed bilateral swap agreements in yuan with foreign central banks; and it has liberalized the use of deposits in yuan in Hong Kong and London. Although internationalization still has a long way to go and will depend on reforms being carried out (see, e.g., Dobson and Masson, 2009; Eichengreen, 2011; Prasad and Ye, 2012; Yu, 2012; Cohen, 2014), it is noteworthy that the first steps were undertaken while the currency was still being carefully managed with respect to the US dollar. According to Vallée (2011), China could move far along the road of internationalization without switching to a free floating exchange-rate regime. However long the transition period, it is difficult to envisage a fully multipolar global economy without a considerable reshaping of the IMS (Angeloni et al., 2011; Bénassy-Quéré and Pisani-Ferry, 2011). Eichengreen (2013) considers that in the next decades, the yuan will likely play at least some international role, akin to that played by the Swiss franc, the yen or the euro today. A multipolar monetary system would alleviate the Triffin dilemma by diversifying the sources of international liquidity. Furthermore, multipolarity could act as a disciplinary device, with investors being offered a choice between several currencies with equivalent liquidity features but issued by countries with varying depths of imbalances (Eichengreen, 2010; Kwan, 2001). Some authors have however argued that a multipolar monetary system could raise exchange-rate volatility since greater substitutability across key currencies would translate into more frequent and larger portfolio reallocations (United Nations, 2009). Extensive literature on the impact of exchange-rate volatility on growth is available. On the theoretical side, Ethier (1973), Clark (1973) and Hooper and Kohlagen (1978) show that currency risks reduce trade volumes if firms are risk-averse regardless of whether hedging is available; and Pindyck (1991) and Demers (1991) argue that the irreversible costs associated with uncertainty reduce investment. However, there have been dissenting views (see Franke, 1991; de Grauwe, 1988 for opposite or ambiguous results). On the empirical side, the link between exchange-rate instability and trade,

1 Authors’ calculation based on IMF, Annual Report on Exchange Arrangements and Exchange Restrictions 2014, and World Bank GDPs in current dollar.

Please cite this article in press as: Agnès Bénassy-Quéré, Yeganeh Forouheshfar, The impact of yuan internationalization on the stability of the international monetary system, Journal of International Money and Finance (2015), doi: 10.1016/ j.jimonfin.2015.05.004

ARTICLE IN PRESS A. Bénassy-Quéré, Y. Forouheshfar / Journal of International Money and Finance ■■ (2015) ■■–■■

3

investment or growth has long been considered ambiguous. However, recent studies that exploit disaggregated data, account for heterogeneity, correct for methodological pitfalls, control for financial development and/or focus on medium to long-term measures of exchange-rate uncertainty (that can less easily be hedged) find a negative impact for exchange-rate variability on trade (see e.g. Taglioni, 2002, Byrne et al., 2008, Héricourt and Poncet, 2013 and the meta-analyses of C´oric´ and Pugh, 2010 and Haile and Pugh, 2013), on foreign direct investment (Cavallari and d’Addona, 2013) and on growth (Aghion et al., 2009). Aside from this literature, it should be mentioned that high exchange-rate volatility tends to fuel discontent within the international community, with accusations of ‘beggar-thyneighbor’ policies or even of currency wars (see Eichengreen, 2013). Some countries may react by raising barriers to trade or capital flows. Less scrutinized by the academic literature, anecdotal evidence nevertheless suggests that this channel could be highly powerful. In this paper, we are interested in the impact of moving from a unipolar to a multipolar monetary system and how this may influence exchange-rate stability. The textbook approach to exchange-rate determination states that the bilateral exchange rate between currencies i and j depends on the fundamentals of countries i and j, as well as on market expectations concerning the future evolution of the bilateral exchange rate, but not on shocks in a third country k: such shocks may affect the exchange rate of currency k against both i and j, leaving the i–j rate unaffected. Still, Fratzscher and Mehl (2008) show that the exchange rates between the euro, the dollar and the yen can be significantly affected by statements made on third currencies. Our model provides theoretical underpinnings for such cross effects and studies whether the internationalization of third currencies may reduce them. Specifically, we study the impact of yuan internationalization within a three-country model (the United States, the Euro area, and China).2The internationalization of the yuan is defined as the introduction of this currency in the portfolios of American and European investors, not only for liquidity purposes (fixed share), but also for investment purposes (risk-return arbitrage with the other two currencies). Hence we are interested in both the means-of-payments and the store-of-value functions of the international currency. In this study, we consider two different scenarios for the Chinese exchangerate regime: free-float and dollar-peg. By comparing these regimes, we also deal with the unit-ofaccount function of international currencies. We use a portfolio-balance model à la Blanchard et al. (2005) (hereinafter BGS). To the extent that domestic investors display a home bias (i.e., a preference for home-currency assets), a transfer of wealth from China to the United States (due to both countries having cumulated imbalances of opposite signs) involves an appreciation of the yuan against the US dollar, because the wealth transfer triggers a rise in the global demand for the yuan and a fall in the global demand for the dollar. BGS then study the implications of China pegging its currency to the dollar through foreign-exchange interventions and capital controls. In such a case, a wealth transfer from the United States to China no longer triggers an appreciation of the yuan against the dollar because the People’s Bank of China buys all the dollars released by impoverished US residents. Hence, the global demand for both dollar and yuandenominated assets remains constant, and the dollar–yuan exchange rate is unaffected. However, the global demand for euro-denominated assets falls because US residents sell their euros while Euro area residents, whose wealth was not directly affected by the shock, fail to buy them: the euro depreciates against both the dollar and the yuan. BGS conclude that China’s dollar-peg tends to maintain a relatively weak euro against the dollar in a context of cumulated US current-account deficits. However they do not consider the possibility of the Chinese currency growing internationally: in their setting, foreign investors are not allowed to hold assets denominated in yuan. We find that allowing the yuan to become an international currency would make exchange rates less vulnerable to international shocks. It would also attenuate the distortions arising from China pegging its currency to the dollar. We believe that a more balanced IMS (through yuan internationalization) could act as a substitute for the difficult coordination of exchange rates, which has largely failed since the Plaza (1985) and Louvre (1987) Accords. Indeed, rebalancing the IMS would further stabilize exchange rates between key currencies.

2 Extending the model to more than three currencies would make it difficult to derive clear-cut results, with little benefit in terms of the basic mechanisms.

Please cite this article in press as: Agnès Bénassy-Quéré, Yeganeh Forouheshfar, The impact of yuan internationalization on the stability of the international monetary system, Journal of International Money and Finance (2015), doi: 10.1016/ j.jimonfin.2015.05.004

ARTICLE IN PRESS 4

A. Bénassy-Quéré, Y. Forouheshfar / Journal of International Money and Finance ■■ (2015) ■■–■■

The remainder of the paper is organized as follows. The model is presented in Section 2. In Section 3, we simulate transitory and permanent trade shocks under two different scenarios for the Chinese exchange-rate regime (free-float or yuan–dollar peg) and its currency internationalization (yuan internationalized or not). Section 4 concludes. 2. The model We consider a model with three countries (the United States, Euro area and China) and either two international currencies (the dollar and the euro) or three (the dollar, the euro and the yuan), with the yuan either being pegged to the dollar or floating freely. As in Blanchard et al. (2005), we choose not to build a fully-fledged macroeconomic model with endogenous prices, although it would allow us to study the impact of various shocks on both nominal and real exchange rates. The reason is that a large number of assumptions would be needed with regard to the exchange-rate pass-through, nominal rigidities and monetary policy, for each of the three countries and under two alternative currency regimes, with results being sensitive to these assumptions. In an environment of low inflation, real exchange-rate variations are driven by nominal exchange rates, and thus we believe this simplification to be benign to our objective of understanding the impact of yuan internationalization on the stability of the IMS. 2.1. Accounting framework The accounting framework is summarized in Table 1. Denoting by i the country of residence (i = U,E,C) and by j the currency of investment (j = U,E,C), Wi represents the wealth of asset holders in country i, Di their holdings in domestic currency and Fji their holdings in foreign currency j. Finally, the asset supply in currency j is denoted by Aj. All asset demands and supplies are expressed in the currency of denomination, whereas wealth is expressed in the home currency of each investor. Using the yuan as the numeraire, SU denotes the nominal exchange rate between the US dollar and the Chinese yuan and SE the nominal exchange rate between the euro and the yuan. A rise in SU (resp. SE) refers to an appreciation of the dollar (resp. of the euro) against the yuan. We present the model with three floating international currencies, before showing what happens when the yuan is not international and/or when it is pegged to the dollar. Each row of Table 1 represents the budget constraint of a country, in its own currency. The columns represent the three market-clearing conditions once the demands have been converted into the same currency. Specifically, the three market-clearing conditions write:

AU = DU + FUE + FUC

(1a)

AE = DE + FEU + FEC

(1b)

AC = DC + FCE + FCU

(1c)

2.2. Portfolio allocation We denote by fji the share of currency j in the portfolio of country i’s residents, and by fi the total share of foreign-currency assets in the portfolio of country i. For instance, for the United States:

Table 1 Accounting framework.

United States (in dollar) Europe (in euro) China (in yuan) Total (asset supply in each currency)

Dollar

Euro

Yuan

Total (wealth)

DU FUESU/SE SU FUC AU

FEUSE/SU DE SE FEC AE

FCU/SU FCE/SE DC AC

WU WE WC –

Please cite this article in press as: Agnès Bénassy-Quéré, Yeganeh Forouheshfar, The impact of yuan internationalization on the stability of the international monetary system, Journal of International Money and Finance (2015), doi: 10.1016/ j.jimonfin.2015.05.004

ARTICLE IN PRESS A. Bénassy-Quéré, Y. Forouheshfar / Journal of International Money and Finance ■■ (2015) ■■–■■

f EU =

S E FEU SU WU

fCU =

5

FCU SU and f U = f EU + fCU WU

Following the portfolio-choice model (see Branson and Henderson, 1985), the share of currency j in the portfolio of country i depends on the expected return differential between the assets denominated in currencies j and i, Rji:

Rij =

1 + r j S ej S j 1 + r i Sie Si

(2)

where rj (resp. ri) denotes the interest rate on assets denominated in currency j (resp. i), Sje (resp. Sie) the expected exchange rate of currency j (resp. i) and Sj (resp. Si) the current exchange rate of currency j (resp. i), with both exchange rates being expressed against the numeraire (here, the yuan, i.e. SC = 1). Currency shares are affected by the expected return differentials defined by Equation (2). More specifically, the share of currency j in the portfolio of country i, fji, depends on two different expected return differentials: the differential between assets denominated in j and i (direct channel) and the differential between assets denominated in j and i′ (i′ ≠ i) (indirect channel). Thus, in order to account for both channels, we define the currency shares as:

f ji = α ij + β ln Rij + γ ln Rij′

(3)

with α ij ≥ 0 and β ≥ γ ≥ 0 (the direct channel should obviously be stronger than the indirect one). For instance, the share of the yuan in the US portfolio depends on the expected return differential between China and the United States, RCU, and on the expected return between China and the Euro area, RCE:

fCU = α CU + β ln RCU + γ ln RCE To keep the model parsimonious, we omit monetary policy: all interest rates are assumed to be constant and equal.3Thus, portfolio shares only depend on expected exchange-rate variations:

⎛ S ej S j ⎞ ⎛ S ej S j ⎞ f ji = α ij + β ln ⎜ e ⎟ + γ ln ⎜ e ⎝ Si Si ⎠ ⎝ Si′ Si′ ⎟⎠

(4)

Accordingly, the share of the yuan in the US portfolio increases when the yuan is expected to appreciate against the dollar (first term, direct channel) or against the euro (second term, indirect channel). The latter effect means that the US investor reallocates its portfolio from the euro to the yuan:

⎛ 1 ⎞ ⎛ 1 ⎞ fCU = α CU + β ln ⎜ e + γ ln ⎜ e ⎝ SU SU ⎟⎠ ⎝ S E S E ⎟⎠ We assume rational expectations which, in the context of a deterministic model, boil down to perfect foresight:

S ej = S j ( +1)

(5)

where S j ( +1) denotes the exchange rate for the next period. In the long term, exchange rates are stable, hence portfolio shares are constant: f ji = α ij .

3

Assuming constant, non-zero interest-rate differentials would lead to similar results.

Please cite this article in press as: Agnès Bénassy-Quéré, Yeganeh Forouheshfar, The impact of yuan internationalization on the stability of the international monetary system, Journal of International Money and Finance (2015), doi: 10.1016/ j.jimonfin.2015.05.004

ARTICLE IN PRESS 6

A. Bénassy-Quéré, Y. Forouheshfar / Journal of International Money and Finance ■■ (2015) ■■–■■

2.3. Net foreign asset positions Denoting by NFAi the net foreign asset position of country i expressed in its own currency, we have:

NFAi = W i − Ai

(6)

NFAs sum up to zero across the three countries:

NFAC = − ( SU NFAU + S E NFA E )

(7)

They accumulate over time due to (i) trade imbalances, (ii) interest payments, and (iii) valuation effects. The stock-flow equations are presented in Appendix A. Here we focus on the trade balance. We assume that the trade balance of each country depends on its bilateral exchange rate vis-à-vis the others (price competitiveness effect) and on the comparative wealth of each country (income effect). Assuming similar parameters in the three countries, the trade balance of the United States TBU, and that of the Euro area TBE, write:

1 ⎛ W C SE E ⎞ ⎤ U ⎡ ⎛S ⎞ TBU = θ ⎜ E − SU ⎟ + η ⎢ −W U + ⎜ + W ⎟⎥+ z ⎝ SU ⎠ ⎝ SU SU ⎠⎦ 2 ⎣

(8a)

1 ⎛ W C SU U ⎞ ⎤ E ⎡ ⎛S ⎞ TB E = θ ⎜ U − S E ⎟ + η ⎢ −W E + ⎜ + W ⎟⎥+ z ⎝ SE ⎠ ⎠⎦ SE 2 ⎝ SE ⎣

(8b)

where θ, η > 0, and zU, zE represent exogenous shocks to the trade balances. This formulation is similar to that of Blanchard et al. (2005), except for the income effect which we add in order to account for asset accumulation in the trade balance. The US (resp. European) trade balance is expressed in dollar (resp. in euro). Accordingly, the Chinese trade balance (in yuan) writes:

TBC = − ( SU TBU + S E TB E )

(8)

2.4. Exchange-rate and currency regimes The model outlined in the previous sub-sections describes the benchmark case where the three currencies have international status in the sense that they are to be found in foreign portfolios. In this framework, we can easily introduce the case where the yuan has not been internationalized by setting to zero the share of US and European wealth denominated in yuan (i.e. fCU = fCE = 0). In the benchmark case, we also assume all three currencies to be floating freely. The two exchange rates (SU and SE) are given by market equilibrium conditions (1a,b), with portfolio choices described in Section 2.2 and wealth accumulation specified in Section 2.3 and Appendix A. Alternatively, if the yuan is pegged to the dollar, then SU becomes exogenous. The peg is maintained through endogenous portfolio allocation by Chinese authorities, who raise the dollar-share of their portfolio whenever they want to avoid their currency appreciating against the dollar. Hence, when the yuan is pegged, the model is solved for the euro–yuan exchange rate, SE, and for the share of the dollar in China’s portfolio, fUC (which no longer follows the behavior described in Section 2.2). 2.5. Calibration The calibration of the model is detailed in Appendix B. We simulate the model around two alternative steady states. In the first one, the three countries are symmetric (same wealth, same parameters, zero NFA), except for the status of the yuan before internationalization, which differs from that of the dollar or the euro. In the alternative asymmetric steady state, China’s wealth is smaller than that of the two other countries; furthermore, the NFAs of the United States and of the Euro area are initially negative (the latter being less so), so that China displays a large and positive NFA; finally, the dollar is predominant in China’s Please cite this article in press as: Agnès Bénassy-Quéré, Yeganeh Forouheshfar, The impact of yuan internationalization on the stability of the international monetary system, Journal of International Money and Finance (2015), doi: 10.1016/ j.jimonfin.2015.05.004

ARTICLE IN PRESS A. Bénassy-Quéré, Y. Forouheshfar / Journal of International Money and Finance ■■ (2015) ■■–■■

7

Fig. 1. Currency shares in global official reserves, in percent of allocated reserves. Source: International Monetary Fund, COFER quarterly data.

portfolio. This second steady state is more realistic than the first one, but the simulations are more difficult to interpret due to several sources of asymmetry. Yet, the impact of the shock on trade balances stays qualitatively the same for these two different equilibriums. In both cases, we set the single interest rate to 2%. At the initial equilibrium, trade is balanced for each country, and exchange rates and NFAs are constant. In the baseline case (symmetric steady state, free floating regime for all countries, yuan internationalized), we set αji = 0.1 which means that, in the absence of expected return differentials, each foreign currency accounts for 10% of a country’s wealth. Hence, the total internationalization of each portfolio is 20%. Alternatively, when the yuan is not an international currency, we set α CU = α CE = 0 , so that the total internationalization of the US and the European portfolios is only 10%. Note that in all cases, there is a strong home bias: absent return differentials, an increase in a country’s wealth will lead to a rise in said country’s demand for all three currencies, yet mostly for the home currency. In all cases, we set β = 0.1 and γ = 0.05. For instance, whenever the yuan is expected to appreciate by 10% against the dollar, the share of the yuan in the US portfolio increases by 1 percentage point (pp); whenever the yuan is expected to appreciate by 10% against the euro, the share of the yuan in the US portfolio increases by 0.5 pp; so if the yuan is expected to appreciate by 10% against both the dollar and the euro, the share of the yuan in the US portfolio rises by 1.5 pp. These orders of magnitude are consistent with observed variations in the currency shares of official reserves (Fig. 1): over the 1999Q1–2014Q1 period, the standard deviation of currency shares was 3 pp for the dollar and the euro, and less than 1 pp for the yen and the pound sterling; during the same period, the coefficient of variation of nominal, bilateral exchange rates between the dollar, the pound, the yen and the euro ranges from 10% (pound/dollar) to 18% (yen/pound), at the quarterly frequency.4 Turning to NFA accumulation, we set θ = 1 and η = 0.1. A 10% appreciation of the dollar against either the yuan or the euro (starting from a unitary exchange rate) reduces the US trade balance by 10, i.e.

4

Authors’ calculations based on ECB quarterly data.

Please cite this article in press as: Agnès Bénassy-Quéré, Yeganeh Forouheshfar, The impact of yuan internationalization on the stability of the international monetary system, Journal of International Money and Finance (2015), doi: 10.1016/ j.jimonfin.2015.05.004

ARTICLE IN PRESS A. Bénassy-Quéré, Y. Forouheshfar / Journal of International Money and Finance ■■ (2015) ■■–■■

8

10% of initial wealth. In turn, a 10% increase in US wealth, all other things being equal, leads the trade balance to fall by 1, i.e. 1% of initial wealth. The sensitivity of the results to the parameters of the trade balance (θ, η) and of portfolio shares (β, γ) is studied in Appendix C. 3. Impact of a trade shock We simulate the impact of an asymmetric trade shock which shifts wealth from the United States to China and Europe. We first study a temporary shock (Section 3.1) and then move to a permanent shock (Section 3.2). We perform the simulations around the symmetric steady state where all three wealth levels are equal, before studying simulations around the asymmetric steady state (Section 3.3).5 3.1. Impact of a temporary trade shock We first study the impact of a temporary trade shock that corresponds to stylized global imbalances: a negative shock to the US trade balance (zU = −2) that is equally distributed between the Euro

Fig. 2. Impact of a temporary trade shock, floating regime. (Deviations from initial, symmetric equilibrium. Source: Author’s calculations.

5

We use DYNARE package on Matlab, see www.dynare.org.

Please cite this article in press as: Agnès Bénassy-Quéré, Yeganeh Forouheshfar, The impact of yuan internationalization on the stability of the international monetary system, Journal of International Money and Finance (2015), doi: 10.1016/ j.jimonfin.2015.05.004

ARTICLE IN PRESS A. Bénassy-Quéré, Y. Forouheshfar / Journal of International Money and Finance ■■ (2015) ■■–■■

9

area (zE = +1) and China (zC = −(zU + zE) = +1 before exchange-rate adjustment). This shock then progressively recedes through a 10% decay per period (zit+1 = 0.9zit). We simulate this temporary shock around the symmetric steady state where wealth levels are identical in the three countries. The shock can thus be interpreted as pp of initial wealths (−2% of US wealth, +1% of European and Chinese wealth). It has no impact on long-term variables which all come back to the initial steady state. Hence, short-term exchange-rates variations can be thought of as misalignments with respect to the long-term equilibrium. a. Floating regime The simulations for the generalized floating regime are reported in Fig. 2. As expected, the US and European NFAs move in opposite directions in the short term: the US NFA falls by twice as much (down by −2 percent of initial wealth) as the European NFA rises (+1 percent of initial wealth), before recovering toward its steady-state value. Since the shock amounts to a wealth transfer from the United States to the other two countries, and given the home bias, the dollar depreciates against both the euro and the yuan. Fig. 2 constrasts the case where all three currencies have equal international status (continuous line) to the case where the yuan is not an international currency (spotted line). While the dynamics of NFAs appears quite similar in both cases, exchange rates respond to a much lesser extent when the yuan is international than when it is not. In the former case, the inflated Chinese demand for yuan-denominated assets is partially matched by reduced demand from the United States. Hence the shock-induced rise in global demand for yuan-denominated assets is more subdued when the yuan is international than when it is not, and the yuan appreciates less against the dollar. The euro also appreciates less against the dollar, although the difference between the two scenarios is more limited for the euro–dollar than for the dollar–yuan. b. Yuan pegged to the dollar We now simulate the same shock when the yuan is pegged to the dollar (Fig. 3). The impact of the shock on the European NFA is similar to that obtained under a free floating regime. In contrast, the US NFA falls by up to 5 percent of US initial wealth, compared to 2 percent in the previous case. This is a direct consequence of not allowing the yuan–dollar exchange rate to stabilize the shock. The euro appreciates against both dollar and yuan, whereas the share of the dollar in China’s portfolio increases endogenously (by up to 3.5 pp) to counter any appreciation of the yuan. The reaction of China’s portfolio allocation is quite similar whether the yuan is international or not. In contrast, the euro appreciates less when the yuan is international compared to when it is not. In the former case, the US investor sells yuan-denominated assets, thus lowering the downward pressure on both the dollar and the dollar-pegged yuan. The impact on the euro is sizable, with a 3.5% appreciation compared with 5.5% when the yuan is not international. It should be remembered here that the trade shock is asymmetric between the United States and Europe, but symmetric between China and the Euro area: the appreciation of the euro against the yuan is only a side effect of China’s peg to the dollar. The simulation shows that this side effect is substantially lower when the yuan is international compared to the case when it is not. Whatever the exchange-rate regime in China, we see that yuan internationalization reduces the reaction of floating currencies following a temporary asymmetric trade shock: medium-term volatility and exchange-rate misalignments are reduced, with little cost in terms of NFA adjustment. Furthermore, when the yuan is pegged to the dollar, asymmetric effects related to the peg are also reduced.

3.2. Impact of a permanent trade shock We now turn to a permanent, asymmetric trade shock. The shock is the same as in the previous section, except that now there is no decay over time. Starting from the symmetric steady state, we have zU = −2 percent of initial US wealth and zE = +1 percent of initial European wealth. The shock is maintained permanently at the same level. Please cite this article in press as: Agnès Bénassy-Quéré, Yeganeh Forouheshfar, The impact of yuan internationalization on the stability of the international monetary system, Journal of International Money and Finance (2015), doi: 10.1016/ j.jimonfin.2015.05.004

ARTICLE IN PRESS 10

A. Bénassy-Quéré, Y. Forouheshfar / Journal of International Money and Finance ■■ (2015) ■■–■■

Fig. 3. Impact of a temporary trade shock, yuan–dollar peg. (Deviations from initial, symmetric equilibrium.) Source: Author’s calculations.

a. Floating regime As in the case of a temporary shock, US and European NFAs move in opposite directions; however, this shift is now permanent. The US NFA stabilizes at −3 percent of initial wealth and the European one at +1.5 percent of initial wealth when the yuan is international (Fig. 4). The dollar depreciates permanently against both currencies, but it depreciates less when the yuan is international. As for a temporary shock, the difference between the two lines is more pronounced for the dollar–yuan than for the euro–dollar. Because of reduced exchange rates responses, NFAs react more to the shock when the yuan is international than when it is not. This reflects the usual trade-off between exchange-rate adjustment and NFA accumulation. Note however that the impact of yuan internationalization is much more pronounced for exchange rates (the long-term dollar depreciation against the yuan is almost halved) than for NFA accumulation (less than 1 pp). b. Yuan pegged to the dollar When the yuan is pegged to the dollar, the euro appreciates against both dollar and yuan in the long term (Fig. 5). As in the case of a temporary shock, the side effect of the peg is more limited for the euro when the yuan is international than when it is not, with a limited impact on NFAs. Please cite this article in press as: Agnès Bénassy-Quéré, Yeganeh Forouheshfar, The impact of yuan internationalization on the stability of the international monetary system, Journal of International Money and Finance (2015), doi: 10.1016/ j.jimonfin.2015.05.004

ARTICLE IN PRESS 0.16

0

0.14

−0.02

0.12

−0.04

0.1

−0.06 S$

Seuro/S$

A. Bénassy-Quéré, Y. Forouheshfar / Journal of International Money and Finance ■■ (2015) ■■–■■

0.08

−0.08

0.06

−0.1

0.04

−0.12

0.02

−0.14

0

0

10

20

30 t

40

50

−0.16

60

11

2

0

10

20

30 t

40

50

60

0

10

20

30 t

40

50

60

0 −0.5

1.5

−1

NFA

E

NFAU

−1.5 1

−2 −2.5

0.5

−3 −3.5

0

0

10

20

30 t

40

50

60

−4

Yuan is internatonalized Yuan is not internationalized

Fig. 4. Impact of a permanent trade shock, floating regime. (Deviations from initial, symmetric equilibrium.) Source: Author’s calculations.

Under the peg regime, the PBoC buys dollar assets in order to avoid an appreciation of the yuan against the dollar. If the yuan is international, the trade shock will induce the US investor to sell its yuan assets, while the Euro area investor will substitute dollars for yuan in his portfolio, reducing the downward pressure on the dollar vis-à-vis the euro. As in the previous case, yuan internationalization has a much greater impact on the exchange rate than on NFA accumulation. In the case of a permanent shock, whatever the exchange-rate regime, we see that yuan internationalization reduces the exchange rate response to the shock while slightly raising NFA accumulation. Again, the asymmetric effect of the yuan peg to the euro–dollar rate is attenuated by yuan internationalization. 3.3. A trade shock around an asymmetric steady state We now perform the same simulations around a more realistic, asymmetric steady state where (i) the United States and, to a lesser extent, the Euro area, have accumulated negative NFAs; (ii) asset supplies in dollar and in euro are larger than asset supplies in yuan; and (iii) the dollar is predominant in China’s portfolio (see Appendix B). Here we focus on a permanent shock which is calibrated in the same way as in Section 3.2. Please cite this article in press as: Agnès Bénassy-Quéré, Yeganeh Forouheshfar, The impact of yuan internationalization on the stability of the international monetary system, Journal of International Money and Finance (2015), doi: 10.1016/ j.jimonfin.2015.05.004

ARTICLE IN PRESS A. Bénassy-Quéré, Y. Forouheshfar / Journal of International Money and Finance ■■ (2015) ■■–■■

12

0.1

0.08 0.07

0.08

0.06 0.05

S

fc

$

euro

0.06

0.04

0.04 0.03 0.02

0.02

0.01 0

0

10

20

30 t

40

50

0

60

2

0

10

20

30 t

40

50

60

0

10

20

30 t

40

50

60

0 −2

1.5

NFA

E

NFAU

−4

1

−6 −8 −10

0.5

−12 0

0

10

20 30

40 t

50

60

−14

Yuan is internationalized Yuan is not internationalized

Fig. 5. Impact of a permanent trade shock, yuan–dollar peg. (Deviations from initial, symmetric equilibrium.) Source: Author’s calculations.

a. Floating regime Under the floating regime, exchange rates respond less than in the symmetric case, while NFAs react more (Fig. 6). This results from two simultaneous effects. First, due to the large dollar-share in China’s portfolio, the dollar depreciation generates a large demand for dollar assets from China in the long term. This stabilizes its exchange rate and slows down the trade balance adjustment (trade adjustment relies relatively more on wealth variations, and relatively less on exchange-rate variations). Second, because the United States has fewer foreign-denominated assets than in the baseline case, revaluation effects are less favorable to this country and its NFA falls sharply even in the short term. On the whole, the results are unchanged qualitatively: with an international yuan, the shock results in a less pronounced appreciation of the euro against the dollar, and in a more pronounced fall of the US NFA. However the impact of yuan internationalization is less pronounced than for the simulations around the symmetric equilibrium. b. Yuan pegged to the dollar In the peg regime, the euro appreciates against both dollar and yuan, but to a lesser extent than in the symmetric case, and the share of the dollar in China’s portfolio increases by less (Fig. 7). When Please cite this article in press as: Agnès Bénassy-Quéré, Yeganeh Forouheshfar, The impact of yuan internationalization on the stability of the international monetary system, Journal of International Money and Finance (2015), doi: 10.1016/ j.jimonfin.2015.05.004

ARTICLE IN PRESS A. Bénassy-Quéré, Y. Forouheshfar / Journal of International Money and Finance ■■ (2015) ■■–■■

0.1

13

0 −0.02

0.08

S$

−0.06

S

euro

/S

$

−0.04 0.06

−0.08

0.04

−0.1 0.02 −0.12 0

0

10

20

30 t

40

50

60

−0.14

3.5

0

10

20

30 t

40

50

60

0

10

20

30 t

40

50

60

0 −1

3

−2 −3

2

NFA

U

NFAE

2.5

1.5

−4 −5 −6

1

−7 0.5 0

−8 0

10

20

30 t

40

50

60

−9

Yuan is internatonalized Yuan is not internationalized Fig. 6. Impact of a permanent trade balance shock, floating regime. (Deviations from initial, asymmetric equilibrium.) Source: Author’s calculations.

the yuan is not an international currency, the sudden appreciation of the euro against the dollar triggers a slight fall in the Euro area’s NFA in the short run due to valuation effects. We conclude that the results obtained from simulations around the symmetric steady state still apply qualitatively when we start from an asymmetric steady state, although to a lesser extent. This was to be expected, given the smaller size of the Chinese economy in this new asymmetric equilibrium. 4. Conclusion A growing literature suggests that the mismatch between a unipolar international monetary system and a multipolar world is not viable in the long run. In addition, China has taken significant steps toward internationalizing its currency since the mid-2000s. In this paper we studied the impact of yuan internationalization on the stability of the IMS. Specifically, based on a three-country, three-currency portfolio model, we studied the impact of asymmetric trade shocks on exchange rates and net foreign asset accumulation, depending on yuan international status. Please cite this article in press as: Agnès Bénassy-Quéré, Yeganeh Forouheshfar, The impact of yuan internationalization on the stability of the international monetary system, Journal of International Money and Finance (2015), doi: 10.1016/ j.jimonfin.2015.05.004

ARTICLE IN PRESS A. Bénassy-Quéré, Y. Forouheshfar / Journal of International Money and Finance ■■ (2015) ■■–■■

0.06

0.05

0.05

0.04

0.04 $

0.06

0.03

fc

Seuro

14

0.03

0.02

0.02

0.01

0.01

0

0

10

20

30 t

40

50

0

60

0

3.5

0

3

−2

2.5

10

20

30 t

40

50

60

10

20

30 t

40

50

60

−4

NFAU

NFAE

2 1.5

−6 −8

1 −10

0.5

−12

0 −0.5

0

10

20

30 t

40

50

60

−14

0

Yuan is internationalized Yuan is not internationalized

Fig. 7. Impact of a permanent trade balance shock, yuan–dollar peg. (Deviations from initial, asymmetric equilibrium.) Source: Author’s calculations.

We find that allowing the yuan to become an international currency would make exchange rates less vulnerable to international shocks. It would also attenuate the distortions arising from China pegging its currency to the dollar. In the case of permanent shocks, greater exchange-rate stability would be ensured along with slightly greater NFA accumulation. In the case of transitory shocks, reduced exchangerate stability can be interpreted as reduced misalignments. We conclude that, after a transition period that may cause monetary instability, yuan internationalization would prove a stabilizing feature of the international monetary system, especially before China moves to a free-floating exchange-rate regime.

Acknowledgement We are grateful to Jean Pisani-Ferry for his contribution at the early stage of this research, to the participants in the CESIfo-CEPII-GEP Conference on China and the World (6–7 September 2012) for useful remarks and to our anonymous referees. All errors remain ours. Please cite this article in press as: Agnès Bénassy-Quéré, Yeganeh Forouheshfar, The impact of yuan internationalization on the stability of the international monetary system, Journal of International Money and Finance (2015), doi: 10.1016/ j.jimonfin.2015.05.004

ARTICLE IN PRESS A. Bénassy-Quéré, Y. Forouheshfar / Journal of International Money and Finance ■■ (2015) ■■–■■

15

Appendix A NFA accumulation To see how NFAs accumulate over time, it is useful to first define them in terms of wealth levels. Here we index all variables for time:

NFAtU = f U WtU −

S E ,t fUE WtE fUC WtC − SU ,t SU ,t

(A1a)

NFAtE = f E WtE −

SU ,t f EU WtU f EC WtC − S E ,t S E ,t

(A1b)

The dynamics of our model is given by NFA accumulation that depends on valuation effects (due to exchange-rate variations), on interest rates (ri, i = U,E,C) and on the trade balance of each country (TBi, i = U,E). Consequently, the NFA position of the United States at the end of period t is equal to the new value of US assets and liabilities of the previous period plus the trade balance of period t:

NFAtU = f EU WtU−1 (1 + r E ) −

fUC SU ,t −1

S S E ,t SU ,t + fYU WtU−1 (1 + r C ) U ,tt −1 S E ,t −1 SU ,t −1 SU ,t

WtC−1 (1 + r U ) − fUE WtE−1 (1 + r U )

S E ,t −1 + TBtU SU ,t −1

(A2a)

By using Equation (A1a) at time t – 1, we can simplify the accumulation of the US NFA as:

⎡ ⎛ S S ⎞ NFAtU = WtU−1 ⎢ f EU ⎜ E ,t U ,t (1 + r E ) − (1 + r U )⎟ ⎝ ⎠ S S E ,t −1 U ,t −1 ⎣ ⎛S ⎞⎤ + fYU ⎜ U ,t −1 (1 + r C ) − (1 + r U )⎟ ⎥ + (1 + r U ) NFAtU−1 + TBtU ⎝ SU ,t ⎠⎦

(A3a)

Similarly, for the Euro area:

S S ⎡ ⎛ ⎞ NFAtE = WtE−1 ⎢ fUE ⎜ (1 + r U ) E ,t U ,t − (1 + r E )⎟ ⎠ S E ,t −1 SU ,t −1 ⎣ ⎝ S ⎛ ⎞ ⎤ + fYE ⎜ (1 + r C ) E ,t −1 − (1 + r E )⎟ ⎥ + (1 + r E ) NFAtE−1 + TBtE ⎝ ⎠⎦ S E ,t

(A4b)

Appendix B Model calibration In the symmetric equilibrium, the three countries are of equal sizes: asset supplies Aj are normalized to 100 (Table B1). When the yuan is international, portfolio shares are symmetric across the different countries and currencies (fij = 0.1), and all NFAs are zero. Alternatively, when the yuan is not international, we have fUC = fEC = 0 while the other portfolio shares remain unaffected. NFAs and exchange rates are adjusted accordingly. In particular, European and US NFAs are slightly negative (less than 10 percent of wealth). This first symmetric equilibrium allows us to highlight possible asymmetries stemming exclusively from the Chinese asymmetric exchange-rate regime (in the peg case) while eliminating asymmetries arising from the initial steady state. This is far from offering a realistic depiction of the Please cite this article in press as: Agnès Bénassy-Quéré, Yeganeh Forouheshfar, The impact of yuan internationalization on the stability of the international monetary system, Journal of International Money and Finance (2015), doi: 10.1016/ j.jimonfin.2015.05.004

ARTICLE IN PRESS 16

A. Bénassy-Quéré, Y. Forouheshfar / Journal of International Money and Finance ■■ (2015) ■■–■■

Table B1 Symmetric steady state with and without yuan internationalization. United States With yuan internationalization Asset supplies (Aj) 100 αEU = αYU = 0.1 Portfolio shares (αji) i NFAs (NFA ) 0 Exchange rates (Sj) 1 Without yuan internationalization Asset supplies (Aj) 100 Portfolio shares (αji) αEU = 0.1; αYU = f YU = 0 NFAs (NFAi) −9.923 1.2596 Exchange rates (Sj)

Euro area

China

100 αUE = αYE = 0.1 0 1

100 αUC = αEC = 0.1 0 -

100 αUE = 0.1; αYE = f YE = 0 -9.923 1.2596

100 αUC = αEC = 0.1 +19.846 -

Source: Authors.

Table B2 Asymmetric steady state with and without yuan internationalization. United States With yuan internationalization Asset supplies (Aj) 110 Portfolio shares (αji) αEU = 0.175; α¥U = 0.025 i NFAs (NFA ) −33.96 Exchange rates (Sj) 1.05428 Without yuan internationalization Asset supplies (Aj) 110 Portfolio shares (αji) αEU = 0.175; αYU = f YE =0 NFAs (NFAi) −36.04 Exchange rates (Sj) 1.006 (flex) or 1.14474 (fix) (exogenous)

Euro area

China

100 αUE = 0.175; αYE = 0.025 −6.32 0.92

50 αUC = 0.4; αEC = 0.1 +40.28 -

100 αUE =0.175; αYE = f YE = 0 −8.67 1.14476 (flex) or 0.92 (fix)

50 αUC = 0.4; αEC = 0.1 +44.71 -

Source: Authors.

real world, given that the United States and the Euro area are of comparable sizes, but China amounts to only about half of the US size, at current exchange rates. Furthermore, the symmetric equilibrium does not account for the large and negative US NFA, nor for the positive Chinese NFA. Finally, it does not account for the large share of the dollar in Chinese reserves. In the asymmetric equilibrium, asset supplies are 110 in the United States, 100 in the Euro area, but only 50 in China, and the dollar represents a large share (40%) of China’s portfolio while the yuan represents a small share (2.5%) of both Euro area and US portfolios. Finally, the NFA of the United States is far more negative than that of the Euro area (Table B2).

Appendix C Sensitivity analysis Here we present simulations around the symmetric equilibrium with different sets of parameters for portfolio choices and for the trade balance. We concentrate on the permanent trade shock.

Portfolio choices Fig. C1 and C2 show the impact of the trade shock when currency shares are more sensitive to expected returns, respectively in the case of a floating regime and in that of a fixed peg. Specifically, we set β = 1 instead of 0.1, and γ = 0.5 instead of 0.05. For the floating regime, the results are similar to Please cite this article in press as: Agnès Bénassy-Quéré, Yeganeh Forouheshfar, The impact of yuan internationalization on the stability of the international monetary system, Journal of International Money and Finance (2015), doi: 10.1016/ j.jimonfin.2015.05.004

ARTICLE IN PRESS 0

0.14

−0.02

0.12

−0.04

0.1

−0.06 S

0.08

−0.1

0.04

−0.12

0.02

−0.14 0

10

20

30 t

40

50

60

−0.16

2

1

1.5

0

1

−1

NFA 0.5

−2

0

−3

−0.5

0

10

20

30 t

40

50

60

17

−0.08

0.06

0

E

$

0.16

NFAU

Seuro/S$

A. Bénassy-Quéré, Y. Forouheshfar / Journal of International Money and Finance ■■ (2015) ■■–■■

−4

0

10

20

30 t

40

50

60

0

10

20

30 t

40

50

60

Yuan is internatonalized Yuan is not internationalized

Fig. C1. Impact of a permanent trade shock, floating regime. (Deviations from initial, symmetric equilibrium, β = 1, γ = 0.5.) Source: Author’s calculations.

those in our baseline simulation, except for larger initial jumps of both exchange rates. As a consequence of the large short-run appreciation of the euro and of the yuan, the US NFA initially rises while that of the Euro area falls (valuation effects). For the peg regime, we also get a larger jump of the euro–dollar exchange rate in the short run. The Euro area’s NFA falls in the short term.

Trade balance We now consider the case of trade being less sensitive to both exchange rates and wealth, which results in exchange rates having a more limited impact on NFAs. Specifically, we set θ = 0.5 instead of 1, and η = 0.05 instead of 0.1. The results are reported in Figs. C3 and C4. They are qualitatively unchanged compared to the baseline case, but the responses of exchange rates and of NFAs are magnified (they are doubled). Since the trade balance is now less reactive to exchange rates and wealth, both variables need to move more before NFA accumulation is stopped. Please cite this article in press as: Agnès Bénassy-Quéré, Yeganeh Forouheshfar, The impact of yuan internationalization on the stability of the international monetary system, Journal of International Money and Finance (2015), doi: 10.1016/ j.jimonfin.2015.05.004

ARTICLE IN PRESS A. Bénassy-Quéré, Y. Forouheshfar / Journal of International Money and Finance ■■ (2015) ■■–■■

18

0.1

0.08

0.06

$

0.06 fc

Seuro

0.08

0.04

0.04 0.02

0.02 0

0

10

20

30 t

40

50

0

60

2

0

10

20

30 t

40

50

60

0

10

20

30 t

40

50

60

0 −2

1.5 1

NFAU

NFAE

−4

0.5

−6 −8 −10

0 −0.5

−12 0

10

20

30 t

40

50

60

−14

Yuan is internationalized Yuan is not internationalized

Fig. C2. Impact of a permanent trade shock, peg. (Deviations from initial, symmetric equilibrium, β = 1, γ = 0.5.) Source: Author’s calculations.

Please cite this article in press as: Agnès Bénassy-Quéré, Yeganeh Forouheshfar, The impact of yuan internationalization on the stability of the international monetary system, Journal of International Money and Finance (2015), doi: 10.1016/ j.jimonfin.2015.05.004

ARTICLE IN PRESS 0

0.3

−0.05

0.25

−0.1

0.2

−0.15 $

0.35

19

S

Seuro/S$

A. Bénassy-Quéré, Y. Forouheshfar / Journal of International Money and Finance ■■ (2015) ■■–■■

0.15

−0.2

0.1

−0.25

0.05

−0.3

0

0

10

20

30 t

40

50

60

0

3.5

−1

3

−2

2.5

−3

0

10

20

30 t

40

50

60

0

10

20

30 t

40

50

60

U

4

NFA

E

NFA

−0.35

2

−4

1.5

−5

1

−6

0.5

−7

0

0

10

20

30 t

40

50

60

−8

Yuan is internatonalized Yuan is not internationalized

Fig. C3. Impact of trade shock, floating regime. (Deviations from initial, symmetric equilibrium, θ = 0.5, η = 0.05.) Source: Author’s calculations.

Please cite this article in press as: Agnès Bénassy-Quéré, Yeganeh Forouheshfar, The impact of yuan internationalization on the stability of the international monetary system, Journal of International Money and Finance (2015), doi: 10.1016/ j.jimonfin.2015.05.004

ARTICLE IN PRESS A. Bénassy-Quéré, Y. Forouheshfar / Journal of International Money and Finance ■■ (2015) ■■–■■

20

Fig. C4. Impact of a permanent trade shock, peg. (Deviations from initial, symmetric equilibrium, θ = 0.5, η = 0.05.)

References Aghion, P., Bacchetta, P., Rancière, R., Rogoff, K., 2009. Exchange rate volatility and productivity growth: the role of financial development. J. Monet. Econ. 56, 494–513. Angeloni, I., Bénassy-Quéré, A., Carton, B., Darvas, Z., Destais, C., Pisani-Ferry, J., et al., 2011. Global currencies for tomorrow: a European perspective, CEPII Research Report 2011-01/Bruegel Blueprint 13. Bénassy-Quéré, A., Pisani-Ferry, J., 2011. What international monetary system for a fast-changing world economy? In: Boorman, J., Icard, A. (Eds.), Reform of the International Monetary System – The Palais-Royal Initiative. SAGE. Blanchard, O., Giavazzi, F., Sa, F., 2005. International investors, the U.S. current account, and the dollar. Brookings Papers on Economic Activity, Economic Studies Program, vol. 36 (2005–1), pp. 1–66. Branson, W., Henderson, D., 1985. The specification and influence of asset markets. In: Jones, R.W., Kenen, P.B. (Eds.), Handbook of International Economics, vol. 2. pp. 750–805. Byrne, J.P., Darby, J., MacDonald, R., 2008. US trade and exchange rate volatility: a real sectoral bilateral analysis. J. Macroecon. 30, 238–259. Cavallari, L., d’Addona, S., 2013. Nominal and real volatility as determinants of FDI. Appl. Econ. 45, 2603–2610. Clark, P.B., 1973. Uncertainty, exchange risk, and the level of International trade. Western Economic Journal 302–313. Cohen, B.J., 2009. Dollar dominance, euro aspirations: recipe for discord? J. Common Mark. Stud. 47, 741–766. Cohen, B.J., 2014. Will history repeat itself? Lessons for the yuan. ADBI Working Paper No. 453. C´oric´, B., Pugh, G., 2010. The effects of exchange rate variability on international trade: a meta-regression analysis. Appl. Econ. 42, 2631–2644. De Grauwe, P., 1988. Exchange rate variability and the slowdown in growth of international trade. International Monetary Fund Staff Paper 35, 63–84. Demers, M., 1991. Investment under uncertainty, irreversibility and the arrival of information over time. Rev. Econ. Stud. 58, 333–350.

Please cite this article in press as: Agnès Bénassy-Quéré, Yeganeh Forouheshfar, The impact of yuan internationalization on the stability of the international monetary system, Journal of International Money and Finance (2015), doi: 10.1016/ j.jimonfin.2015.05.004

ARTICLE IN PRESS A. Bénassy-Quéré, Y. Forouheshfar / Journal of International Money and Finance ■■ (2015) ■■–■■

21

Dobson, W., Masson, P., 2009. Will the yuan become a world currency? China Econ. Rev. 20, 124–135. Eichengreen, B., 1987. Hegemonic stability theories of the international monetary system. Working Paper 2193, NBER. Eichengreen, B., 2010. Managing a multiple reserve currency world. Mimeo, April 2010. Eichengreen, B., 2011. Exorbitant Privilege. Oxford University Press. Eichengreen, B., 2013. Number one country, number one currency? World Econ. 36 (4), 363–374. Ethier, W., 1973. International trade and the forward exchange market. Am. Econ. Rev. 63 (3), 494–503. European Central Bank, 2014. The international role of the euro. Annual report, July. Fahri, E., Gourinchas, P.-O., Rey, H., 2011. Reforming the International Monetary System. Mimeo, March. Franke, G., 1991. Exchange rate volatility and international trading strategy. Journal of International Money and Finance 10, 292–307. Fratzscher, M., Mehl, A., 2008. Do China and oil exporters influence major currency configurations? ECB Working Paper No. 973, December. Haile, M.G., Pugh, G., 2013. The effects of exchange rate variability on international trade: a meta-regression analysis. J. Int. Trade Econ. Dev. 22 (3), 321–350. Héricourt, J., Poncet, S., 2013. Exchange rate volatility, financial constraints, and trade: empirical evidence from Chinese firms. World Bank Econ. Rev. 1–29. Hooper, P., Kohlagen, S.W., 1978. The effects of exchange rate uncertainty on the prices and volume of international trade. J. Int. Econ. 8, 483–511. Ivashina, V., Scharfstein, D., Stein, J., 2012. Dollar funding and the lending behavior of global banks. NBER Working Paper, No. 18520, November. Kindleberger, C., 1973. The World in Depression. University of California Press. Kindleberger, C., 1981. Dominance and leadership in the international economy. Int. Stud. Q. 25 (2), 242–254. Kwan, C.H., 2001. Yen Bloc: Towards Economic Integration in Asia. Brookings Institution Press, Washington, D.C. Mateos y Lago, I., Duttagupta, R., Goyal, R., 2009. The debate on the international monetary system. Staff Position Note 09/26, International Monetary Fund. Pindyck, R., 1991. Irreversibility, uncertainy, and investment. J. Econ. Lit. 29, 1110–1148. Prasad, E., Ye, L., 2012. Will the yuan rule? Finance & Development, March, pp. 26–29. Taglioni, D., 2002. Exchange rate volatility as a barrier to trade: new methodologies and recent evidence. Econ. Int. 89–90, 227–259. Triffin, R., 1960. Gold and the Dollar Crisis: The Future of Convertibility. Yale University Press. United Nations, 2009. Report of the Commission of Experts of the President of the United Nations General Assembly on Reforms of the International Monetary and Financial System, June. Vallée, S., 2011. The internationalisation path of the RMB. Bruegel Working Paper, 2011-05, March. Yu, Y., 2012. Revisiting the internationalization of the yuan. ADBI Working Paper No. 366, July. Zhou, X., 2009. Reform of the international monetary system. Essay posted on the website of the People’s Bank of China, 9 April.

Please cite this article in press as: Agnès Bénassy-Quéré, Yeganeh Forouheshfar, The impact of yuan internationalization on the stability of the international monetary system, Journal of International Money and Finance (2015), doi: 10.1016/ j.jimonfin.2015.05.004