The “real” causes of China’s trade surplus

Kjetil Storesletten, Zheng Song, Fabrizio Zilibotti 02 May 2010



Over the last two decades, China has run large trade surpluses. Its foreign reserves swelled from $21 billion in 1992 (5% of its annual GDP) to $2.4 trillion in June 2009 (close to 50% of its GDP). The effect of this gigantic build up of reserves has been a source of growing public attention in the context of the debate on global imbalances. This debate has gained momentum during the global crisis. Lobbyists and politicians voice the popular concern that by swamping western markets with its products, China contributes to the failure of domestic firms and job losses. The call for protectionism is mounting.

Did China engineer a trade surplus?

A common argument, especially in the US, is that the culprit of global imbalances is the exchange rate manipulation carried out by the Chinese authorities, who peg the renminbi to the dollar at a low value. According to Fred Bergsten, head of the Peterson Institute for International Economics, the renminbi is undervalued by at least 25% to 40%. This "hostile" policy raises calls for robust retaliation.

While economists have so far opposed any measure that might ignite a vicious cycle of trade retaliations and protectionism, even their front is cracking. Krugman (2010) advocated using the threat of a 25% import surcharge to force China to revalue. Last month, 130 lawmakers signed a letter asking the US Treasury to increase tariffs on Chinese-made imports. On 12 April 2010, Barack Obama openly criticised the Chinese exchange-rate policy in front of Hu Jintao, arguing that currencies should "roughly" track the market so that no country has an advantage in trade. Meanwhile, Senator Charles Schumer called for high tariffs against Chinese imports in order to force Beijing to revalue its currency.

The exchange rate manipulation premise

The manipulation thesis rests on the simple postulate that the imbalance itself is evidence of a misalignment of the exchange rate. Letting market forces determine the exchange rate would restore trade balance.

This argument has weak foundations. What matters is the real exchange rate, not the nominal one. While the Chinese surplus has persisted for almost two decades, the real exchange rate has remained as flat as a pancake (see McKinnon 2006, Figure 3). A misaligned real exchange rate should feed domestic inflation, e.g., by increasing the demand of non-traded goods and stimulating domestic wage pressure. Yet, until very recently it does not appear as if China has experienced any major inflationary pressure – between 1997 and 2007 the inflation rate was on average about the same as in the US. Moreover, wages have grown slower than output per worker (see Banister 2007).

In a recent article on this site, Helmut Reisen (2010) shows that a large part of the alleged undervaluation of the renminbi can be attributed to the Balassa-Samuelson effect (i.e., the fact that non-traded goods do not follow the law of one price and are relatively cheap in developing countries). He concludes that "the undervaluation in 2008 of the renminbi was only 12% against the regression-fitted value for China's income level." This is by no means a large number: “Both India and South Africa (which had a current-account deficit) were more undervalued in 2008.” In summary, while it is reasonable to expect some appreciation of the real exchange rate in the years to come (through either inflation or adjustments in the nominal exchange rate), the government manipulation of the nominal exchange rate is unlikely to be the primary cause of the two-decades-long imbalance.

A “real” explanation: Growing like China

What, then, can account for the Chinese surplus? We believe that the answer lies in real (i.e., structural) factors rather than in nominal rigidities. Let us look at the imbalance from an asset flow perspective:
A trade surplus implies a net capital outflow. At first view, it is puzzling that China is a net capital exporter:

  • Aggregate and firm-level data show that the rate of return on investments is much higher in China than in OECD countries (see Bai et al. 2006; Song et al. forthcoming); and
  • Total factor productivity in Chinese manufacturing has been growing at a startling 6% annual rate in the last decade (Bosworth and Collins 2008; Brandt et al. 2009).

So why would Chinese savings be invested in low-yielding US government bonds instead of reaping the high returns of domestic investments?

In a forthcoming article in the American Economic Review titled “Growing like China” (Song, Storesletten, and Zilibotti), we propose an answer that relies on a simple economic mechanism without resorting to any exchange rate manipulation story. The predictions of our theory are consistent both qualitatively and quantitatively with the stylised facts of economic growth in China since 1992, i.e. high output growth, sustained returns on capital investment, an extensive reallocation within the manufacturing sector, and a falling labour share.

The building blocks of the theory are:

  • Differences in productivity across firms (some, mostly privately-owned, firms use more productive technologies;
  • Asymmetric financial imperfections (less productive, mostly state-owned, firms survive because of better access to credit markets).

Due to legal and financial market imperfections, private firms are financing themselves out of internal savings. The rapid growth of these self-financed firms and corresponding down-sizing of bank-financed firms creates and artificial lack of domestic investment opportunities for Chinese banks. As a consequence, a growing share of domestic savings is invested abroad and this generates a capital account deficit and matching current account surplus.

The theory rests on two sets of observations:

First, since 1992, China's urban sector has experienced a rapid transition from state-owned to private enterprises (see Figure 2 in SSZ). For example:

  • Using a firm-level data set including all firms with sales over 5 million yuan, the share of private employment in manufacturing increased from 4% in 1998 to 56% in 2007 (where the private share is defined as the ratio of employment in domestically owned private enterprises over employment in state-owned and private enterprises).
  • Aggregate data on urban employment indicate a similar trend, albeit more gradual, from 7% private employment in 1992 to 54% in 2007. The absolute numbers are large; private employment in Chinese cities increased by at least 129 million workers between 1992 and 2007 – equivalent to the entire US employment.

Since 1992, state-owned enterprises must play by market rules and can be shut down unless they are profitable. Moreover, since 1997, the ruling Communist Party has officially endorsed an increase in the role of private firms in the economy. In short, the main reason for this rapid reallocation is that private firms are more productive and more profitable than state-owned firms (see also Dollar and Wei 2005).

Second, private firms are more financially constrained than state-owned firms (Boyreau-Debray and Wei 2005).
In our paper, we document that state-owned firms finance investments through bank or equity finance more than private domestically-owned firms do – by a factor of three. The latter firms instead rely heavily on retained earnings, personal savings of entrepreneurs, and informal loans from family and friends.

The reason for this discrimination in credit markets is twofold:

  • Banks are state owned and favour state-owned firms;
  • The weakness of the legal system makes it hard for banks to have loans repaid.

There are other symptoms from private firms being credit constrained. For example, state-owned firms have substantially more capital per worker than private firms.

Putting these two observations together suggest an explanation for China’s capital exports. China moved from a situation in the early 1990s where banks were lending substantial funds to the mostly state-owned manufacturing sector to a situation where bank loans to domestic firms have shrunk dramatically. The reason for the shrinkage is the downsizing of the banks’ main borrowers (state-owned firms) teamed with legal and political problems that stymie bank lending to privately-owned firms even thought these private firms are credit constrained.

Just as this constellation of problems narrowed Chinese banks’ domestic lending opportunities, the supply of deposits boomed. During the same period, households increased their savings deposited with banks. Chinese banks are awash with cash and it is this excess of funds that has fed the purchase of foreign bonds (through the central bank that formally has the monopoly on foreign assets).

Figure 1 below shows that the increasing difference between domestic bank deposits and domestic bank loans (dotted red line) very closely tracks the increasing foreign surplus (solid blue line).

Figure 1. China’s foreign surplus and deposit-loans gap of Chinese banks

Source: China Statistical Yearbook, various issues.

Our theory is supported by two important facts.

  • First, the timing of structural change from state-owned to private enterprises closely follows that of the accumulation of foreign reserves.
  • Second, the breakdown of the net surplus (savings minus investments) across 31 Chinese provinces suggests the same pattern in the cross section. The net surplus is systematically larger in provinces with a larger increase in the private employment share (see Table 1 in SSZ).

Outside of China: South Korea and Taiwan

The Chinese build-up of foreign reserves is so spectacular because of the size of the country and its global implications. However, the coexistence of high productivity growth and trade surplus has been documented as a regularity that goes beyond the Chinese case (see Gourinchas and Jeanne 2007). Our paper shows that the reallocation from less to more financially constrained firms can explain salient aspects of the experiences of Korea and Taiwan. In particular, after 1980, both countries experienced accelerating productivity growth combined with a large balance of payment surplus.

In the early 1980s, during the initial stage of its industrialisation process, Korea relied substantially on foreign loans and had one of the highest ratios of foreign debt to GDP among developing countries. This imbalance was significantly reduced in the second half of the 1980s when Korea saw booming growth and a series of large current account surpluses. This turning point coincided with a change in the Korean development strategy, which had earlier on relied on the strong integration between banks and large firms (chaebol). In the 1980s, reallocation from chaebol to small enterprises became an important driver of the growth process. The number of small enterprises more than doubled between 1980 and 1990, and their employment share in manufacturing increased accordingly, a trend that continued in the early 1990s.

Similarly, Taiwan experienced large trade surpluses in the 1980s. In the same period, the employment share of firms employing less than 100 persons (which were more credit constrained) increased from 39% in 1976 to 59% in 1991.

Both Korea and Taiwan, while not coming from a communist past, experienced a significant reallocation within the manufacturing sector characterised by a strong growth of credit-constrained high-productivity firms. Reallocation came hand in hand with an accelerating productivity growth and balance of payment surpluses. These features are common with the Chinese experience.


Our theory challenges the popular view that trade surplus is artificially engineered by the Chinese authorities. We argue instead that the surplus arises from structural factors.

What are the policy implications?

  • First, the West should push and help China improve the efficiency of its credit system so as to channel more of the growing domestic savings toward high-return private investments.
  • Such reform would yield a double dividend. It would reduce the foreign surplus while increasing both wage growth and the rate of return on domestic savings, in principle allowing China to sustain the same stellar growth rates with lower savings.
  • Second, part of the reason behind the large surplus is the high savings rate of China.
  • Why Chinese households save so much is not fully clear, but part of these savings may be precautionary (as argued by Mendoza et al. 2009) and related to the low safety net of families, or the gender imbalance (Wei 2010). Introducing welfare state elements could improve the lives of Chinese people and increase their consumption. The government could easily finance welfare state policies out of the surplus that is currently investing in low-yield bonds.

The final policy implication is obvious – the call for trade sanctions against China may be unwarranted as well as dangerous.

Disclaimer: The opinions expressed here are those of the authors and not necessarily those of the Federal Reserve Bank of Minneapolis or the Federal Reserve System.


Bai, Chong-En, Chang-Tai Hsieh, and Yingyi Qian (2006), “The Return to Capital in China”, Brooking Papers on Economic Activity 2, 61–88.
Banister, Judith (2007), “Manufacturing in China Today: Employment and Labour Compensation”, Economics Program Working Paper 07-01, The Conference Board.
Bosworth, Barry and Susan M Collins (2008), “Accounting for Growth: Comparing China and India”, Journal of Economic Perspectives 22(1):45–66.
Boyreau-Debray, Genevieve, and Shang-Jin Wei (2005), “Pitfalls of a State-Dominated Financial System: The Case of China”, NBER Working Paper 11214.
Brandt, Loren, Johannes Van Biesebroeck, and Yifan Zhang (2009), “Creative Accounting or Creative Destruction? Firm-level Productivity Growth in Chinese Manufacturing”, NBER Working Paper 15152.
Dollar, David and Shang-Jin Wei (2007), “Das (Wasted) Kapital: Firm Ownership and Investment Efficiency in China”, NBER Working Paper 13103.
Gourinchas, Pierre-Olivier, and Olivier Jeanne (2007), “Capital Flows to Developing Countries: The Allocation Puzzle”, NBER Working Paper 13602.
Krugman, Paul (2010), “Taking On China”, The New York Times, 14 March.
McKinnon, Ronald (2006), "China's Exchange Rate Trap: Japan Redux?”, American Economic Review, 96(2), 427–431.
Mendoza, Enrique G, Vincenzo Quadrini, and Jose-Víctor Ríos-Rull (2009), “Financial Integration, Financial Development, and Global Imbalances”, Journal of Political Economy, 117(3), 371–416.
Reisen, Helmut (2010), “Is China's Currency Undervalued?”,, 16 April
Song, Zheng, Kjetil Storesletten, and Fabrizio Zilibotti. Forthcoming, “Growing like China”, American Economic Review.
Wei, Shang-Jin (2010), “The mystery of Chinese savings”,, 6 February.




Topics:  Global crisis International trade

Tags:  exchange-rate policy, Global imbalance, savings-investment imbalances

Professor, University of Oslo; Research Fellow, CEPR

Associate Professor of Business at the Booth School of Business, University of Chicago

Tuntex Professor of International and Development Economics, Yale University; CEPR Research Fellow