In the recent US presidential campaign, China was accused again of currency manipulation. In other words, the Chinese central bank is accused of maintaining the exchange rate at an artificially low level compared to its equilibrium value, including heavy intervention in the foreign exchange market. There has been a fierce debate on this issue in recent years, including on VoxEU.org (e.g., Persaud 2011, Reisen 2011, Reisen et al. 2011, Storesletten et al. 2010).
Underlying this debate is the difficult question of determining the equilibrium value of the renminbi. In recent years, the degree of undervaluation has arguably become a less serious issue since the renminbi has significantly appreciated in real terms as Figure 1 illustrates.
Figure 1. Real effective exchange rates
Source: International Monetary Fund; Based on Consumer Price Index.
Besides measurement issues in the equilibrium currency value, several economists have objected to the ‘currency manipulation’ perspective on more conceptual grounds. For example, Song et al. (2010) argue that the accumulation of foreign reserves reflects the country’s net saving, caused in particular by financial market imperfections. With a related perspective, in a recent paper we examine the optimal exchange rate policy in a model economy that shares features with the Chinese economy. We consider a semi-open economy, where the private sector faces capital controls while the central bank has free access to international capital markets. We also assume that the economy faces tight credit constraints and grows at a fast rate. In such a setting, we show that the optimal policy is to initially depreciate the currency and accumulate reserves, and to gradually appreciate the exchange rate later on. This policy allows the central bank to provide saving instruments to economic agents through its liabilities, which helps them accumulate assets and overcome credit constraints. Our results suggest that China should keep on appreciating its currency as the need for additional saving instruments gradually subsides. This medium-term appreciation coming from a gradual decline in saving rates can be contrasted with the so-called ‘Balassa-Samuelson effect’, where the gradual appreciation comes from productivity differentials.
Capital controls, reserve accumulation and the real exchange rate
Economists usually consider that the equilibrium real exchange rate is related to domestic saving and investment. Other things being equal, higher desired saving (or lower desired investment) should reduce demand and pressure on domestic prices, implying a real depreciation. This can also be seen from the perspective of the current account, which equals the difference between domestic saving and investment. Higher domestic saving implies a larger current account surplus, which requires a depreciated real exchange rate if it is to materialise at full employment.
This has led some commentators to argue that China could not manipulate its exchange rate in the long term. An undervalued nominal exchange rate ought to boost exports, overheat the economy, and result in higher domestic prices. This would bring the real exchange rate back to an equilibrium value consistent with desired saving and investment.
However, in a semi-open economy with strict restrictions on private capital flows, the current account balance is essentially given by the increase in foreign reserves, which the central bank controls. But how can the central bank affect domestic saving and investment? The answer lies in the central bank balance sheet: the counterpart of foreign reserves is domestic liabilities, for instance sterilisation bonds, or accounts of commercial banks. A central bank which buys international reserves and finance them by issuing domestic saving instruments can force the private sector to increase its net saving. Adjustment comes from the domestic interest rate, which can deviate from the world interest rate because of the capital controls. Therefore, with capital controls, a policy of reserve accumulation can determine the equilibrium real exchange rate by directly affecting the underlying balance of saving and investment. Jeanne (2012) provides a formalisation of this argument.
A benchmark for ‘exchange rate manipulation’
Can we talk of ‘exchange rate manipulation‘? Doing so would require comparing the actual accumulation of reserves, current account, and real exchange rate to some ‘normal’ levels. A possible benchmark is the open economy with unrestricted private capital flows, where private saving could quite conceivably be very large. It is possible that the accumulation of reserves just reflects a strong need for saving by the private sector. In that case, the central bank would just serve as intermediary between the private sector and the international capital market. We might then find it difficult to talk about currency manipulation.
But reproducing the aggregate net saving of an open economy is not necessarily optimal. With strong financial frictions, policy intervention might improve on the free market allocation. As explained above, reserves and exchange rate policy imply changes in the provision of saving instruments to the private sector. In an economy with strong financial frictions, providing the right amount of saving instruments is an important feature of an optimal policy.
A simple dynamic model with ‘excess’ saving
We consider a model where economic agents can be either borrowers or lenders. Lenders save part of their income today to be used in future times when their income falls short of their desired expenditures. Borrowers use their accumulated assets and borrow to finance their desired expenditures. We assume that borrowing is limited. This provides an incentive for lenders to save more, since they know they will not be able to borrow a lot when they need it. However, there are not many saving instruments, precisely because borrowing is limited.
In such a setting, the central bank can help lenders accumulate assets and overcome their borrowing constraints in bad times by providing saving instruments for them to buy in good times. The central bank then uses the proceeds to buy foreign reserves. Since this policy increases domestic net saving today, it also initially depreciates the real exchange rate. In the future, the economy is richer since it has larger net foreign assets. Aggregate spending can then be higher and the real exchange rate appreciates in the future.
One experiment we consider is an increase in the growth rate. We assume that growth jumps to 10%, but then gradually declines as on a convergence path. We examine the dynamics of the optimal policy in this case. Our results are shown in Figure 2.
Figure 2. Real exchange rate (left) and international reserves (right)
Note: Percentage change from initial value.
The left-hand picture shows the optimal evolution of the real exchange rate. We see that the initial reaction is depreciation. The reason for this is that the increased growth rate tightens borrowing constraints and generates a higher saving need, which is accommodated by the central bank policy. This is reflected in the right-hand picture, which shows the optimal accumulation of reserves. Given the lack of private capital mobility, changes in reserves are equal to the current account surplus. If we looked at the longer run, we would see that the real exchange appreciates to a higher level than its initial value and that the level of international reserves stabilises.
A weak currency value does not necessarily reflect currency manipulation. A major finding of our analysis is that a fast growing economy with strong financial frictions and a high saving rate naturally has a weak currency. In the absence of private capital mobility, it is then optimal for the central bank to accumulate international reserves to accommodate the high saving rate. However, in the medium term this weakness should disappear if saving gradually declines. Instead of focusing on currency manipulation, it is therefore more useful to insist on the right policies to foster Chinese consumption. This will naturally lead the renminbi to its appreciating path.
Bacchetta, P, K Benhima, and Y Kalantzis (2012), "Optimal Exchange Rate Policy in a Growing Semi-Open Economy," mimeo.
Jeanne, O (2012), "Capital Account Policies and the Real Exchange Rate," NBER Working Paper, 18404.
Persaud, A (2011), “The fog of currency war”, VoxEU.org, 16 February.
Reisen, H (2011), “The renminbi and poor-country growth”, VoxEU.org, 5 December.
Reisen, H, Moritz Schularick and Edouard Turkisch (2011), “Half a century of large currency appreciations: Did they reduce imbalances and output?”, VoxEU.org, 2 March.
Storesletten, Kjetil, Zheng Song and Fabrizio Zilibotti (2010), “The “real” causes of China’s trade surplus”, VoxEU.org, 2 May.
Song, Z M, K Storesletten and F Zilibotti (2010), ”Growing like China,” American Economic Review, 101, 202-241.