STRATEGY
China's Rush for External Assets
August 9, 2007
James McCormack, Fitch Ratings

With China's official foreign-exchange reserves exceeding US$1 trillion, there is considerable interest in how they are invested. Given the magnitude of Chinese holdings of US government securities, any discernable move towards diversification could have a significant impact on global fixed-income and currency markets. Moreover, it was announced on the sidelines of the recent National People's Congress that a new agency would be established to more actively manage a portion of the reserves, and report directly to the State Council. This has focused attention again on how China's official reserves are invested, and raised questions as to possible changes in asset allocation by the new agency.

The Growth of China's Reserves

Changes in the level of official foreign-exchange reserves are typically the result of decisions by the authorities managing monetary and exchange rate policy. A reduction in reserves, for example, would follow a central bank's purchase of local currency in the foreign-exchange market in order to prevent a disorderly depreciation. Conversely, the prevention of a disorderly appreciation would require the central bank to sell local currency and accumulate foreign-exchange reserves.

Official reserves can also fund external debt repayments, though this would normally be for public sector entities, and the decision to draw on reserves (as opposed to purchasing foreign currency in the market) may, in any event, be motivated ultimately by exchange rate considerations. More unusually - but as has been the case in China - reserves can be used for public policy initiatives, such as the recapitalisation of state-owned banks. A more common public policy approach to the use of external assets is the establishment of stabilisation funds by commodity exporters.

In the case of China, official reserves have increased due to current account surpluses, large capital inflows, limited capital outflows and an exchange rate policy that allows for only gradual appreciation. China's reserves surpassed US$100bn in 1996 and US$200bn in 2001. The average annual increase over this period was US$21bn - now an amount which monthly reserve additions regularly exceed. China's reserves moved above those of Japan (AA) in February 2006, and are now higher by about US$170bn.

From 2001 to 2004, official reserve accumulation accelerated rapidly. Up until 2004, the balance of payments was driven by net capital inflows, mostly equity FDI. In 2005 and 2006, current account surpluses of US$161bn and an estimated US$208bn, respectively, were considerably larger than net capital inflows, as the merchandise trade surplus came to dominate the balance of payments. Fitch expects this to continue in 2007.

The rapid increase in the trade surplus has become a focus of international policymakers, especially in the US and European Union, with which China maintains its largest surpluses. US and European officials have repeatedly pointed out that China's exchange rate policy unfairly supports export competitiveness by limiting the appreciation of the renminbi. In addition to the trade balance, they cite official reserve accumulation as further evidence the currency is undervalued, since the PBOC must, by definition, be intervening in the market. This argument, however, discounts the role of China's controlled capital account. If the capital account were completely open, it is conceivable that current account surpluses could be matched by capital outflows, thus alleviating upward pressure on the exchange rate. Capital account liberalisation is taking place, but at a measured pace. It is likely that continued current account surpluses will be accompanied by large foreign-exchange reserve accumulation.

Even with only partial capital account liberalisation, net debt flows were negative in 2005, and are estimated to have remained negative last year. This is attributable to net lending abroad - accumulating debt securities as well as the provision of trade credits and other loans to non-residents. In the absence of these outflows, China's reserves would have grown by nearly US$300bn in 2005, and more than US$300bn in 2006.

China's Total External Assets

Capital outflows from China may be partially restricted, but the country has managed to build a sizeable external asset position. In addition to the net debt outflows noted above, there has also been some outward direct investment and other capital outflows.

The latest available international investment position (IIP) is 2005. China's IIP is rather rudimentary, as it does not allocate assets and liabilities into those held and incurred by different sectors of the economy, but it does provide broad measures of stocks to match up to capital flows in the balance of payments. Fitch has derived the IIP for 2006 based on the agency's balance of payments estimates. At end-2006, Fitch estimates China's external assets amounted to US$1.55 trillion, an increase of US$342bn over the previous year, US$245bn of which was accounted for by official reserves.

At end-2006, official reserves represented 69% of total external assets, with their share having increased by one percentage point in both 2005 and 2006. It is widely recognised that reserves are growing quickly, but other assets are clearly growing nearly as quickly.

Fitch believes most of the estimated US$487bn in non-reserve external assets are held by Chinese banks. These will include portfolio investments abroad as well as deposits and other loans.

According to the banking survey in the most recent IMF International Financial Statistics, banks held external assets of US$272.3bn in December 2006. Statistics from the latest BIS Quarterly Review show total Chinese deposits in banks in BIS member states amounted to US$132bn at September 2006, US$100bn of which were deposits of Chinese banks.

In addition to the outflows, a number of Chinese banks have been listed in the Hong Kong Exchange over the last two years. China Construction Bank, Industrial and Commercial Bank of China (ICBC), Bank of China and Bank of Communications are all listed in Hong Kong. The initial public offering (IPO) of ICBC shares in October 2006 was the world's largest to date, at US$21.9bn. Fitch understands that not all of the funds raised in Hong Kong IPOs were repatriated to the Mainland.

Amounts that remain outside the Mainland would add to banks' external asset positions. The external assets of the non-bank sector are much more difficult to track. Deposits in BIS banks were US$32bn in September 2006. This seems plausible given the cumulative non-bank outflow in the 'other investment asset' category of the financial account of the balance of payments between 2000 and 2005 (latest available) of US$23bn.

China Compared to Other Large Net Creditors

In calculating a country's net external debt, Fitch makes several adjustments to IIP data. First, external portfolio equity holdings and equity-financed FDI assets are excluded from all countries' gross external assets. This ensures the agency's gross external debt figures (which exclude equity liabilities) correspond with gross external asset data. The second adjustment applies only to emerging markets, where Fitch also excludes non-bank private sector external assets, leaving only banks' external assets and those of the official sector (foreign-exchange reserves). In Fitch's judgment, this provides a better measure of liquid external assets that would be available if an emerging market faced an external financing crisis.

Private non-bank external assets may or may not be repatriated in a crisis, and it would certainly be more difficult for the sovereign to enforce repatriation. The above adjustments to China's external assets as reported and estimated in the IIP do not make a material difference. Instead of gross external assets of US$1.55 trillion, China's assets for the purpose of calculating its net external debt are estimated to have been US$1.47 trillion at end-2006. There are no recorded external portfolio equity holdings, and outward FDI stocks (assuming 100% equity financing) are modest at US$79.5bn.

There were 13 rated sovereigns with gross external assets of more than US$1 trillion at end-2006. Nine of these sovereigns are rated AAA, the exceptions being China (A), Belgium (AA+), Italy (AA-) and Japan (AA). At US$1.47 trillion, China's stock of external assets was the ninth largest in the world last year.

China is clearly not unique in having amassed a large external asset pool. Its assets are unusual, however, in that they are dominated by official foreign-exchange reserves. Advanced economies, and particularly those in the euro area, tend to have low levels of reserves. As noted above, China's reserves have been the highest in the world since February 2006.

The concentration of external assets in the public sector is an important supporting point from a sovereign rating perspective, especially in an emerging market where non-bank private-sector external assets are not taken into account in Fitch's calculation of the country's net external debt.

Official reserves account for more than 70% of China's non-equity external assets. In Japan, reserves are about one-quarter of gross external assets (being an advanced economy, non-bank private-sector assets of about US$1.6 trillion are included), but this is much higher than is the case in other large net creditors, indicating China's position as an outlier in this respect. In the UK, with the world's largest external assets of US$7.05 trillion, official reserves were only US$47bn at end-2006, less than 1% of gross external assets.

The concentration of China's external assets in official reserves is consistent with its capital account restrictions and relatively inflexible exchange rate. With an open capital account and fully flexible exchange rate, current account surpluses would result in capital outflows and/or exchange rate appreciation, the latter presumably resulting in lower current account surpluses over time. Neither of these options is exercised fully in China. Capital outflows are restricted and currency appreciation is carefully managed, resulting in the accumulation of official reserves.

The Management of Reserves

The management of China's foreign-exchange reserves is the responsibility of SAFE. It also supervises foreign-exchange flows in the banking system and balance of payments flows in the capital and financial accounts. SAFE does not report directly to the State Council (unlike the PBOC, all government ministries and various other state administrations), as it is organised under the PBOC. The PBOC balance sheet records foreign-exchange assets, as opposed to deposits with SAFE, leaving little doubt as to where ultimate control resides.

The Central Huijin Investment Company is also organised under the PBOC and was established to manage the state's ownership stakes in the banking system. To increase state banks' capital, it injected US$45bn in foreign-exchange reserves into the Bank of China (US$22.5bn) and China Construction Bank (US$22.5bn) in 2003, and another US$15bn into the Industrial and Commercial Bank of China in 2005. These funds were transferred from SAFE to Central Huijin, and official foreign-exchange reserves were reduced accordingly.

The use of reserves in state bank recapitalisations suggests officials have recognised that reserves can - and perhaps should - be managed in different ways. This, of course, is a reflection of the magnitude of the stock of reserves, since China is now well endowed in this regard by any objective measure. It has recently been confirmed publicly by several senior officials that a new agency will be established to manage at least a portion of the foreign-exchange reserves. The new agency is to be elevated in status, reporting directly to the State Council, as do government ministries, the PBOC and the National Development and Reform Commission. Central Huijin will be part of the new agency, possibly with an expanded mandate in terms of domestic investment and additional resources to match its new role. Senior officials have made reference to Singapore's Temasek Holdings, which is 100% government-owned and has large domestic and international investments, as a possible model for China. This would be a significant change, with some of the foreign reserves more actively managed than they have been hitherto, essentially diversifying China's official external holdings to include, for example, equity stakes in foreign enterprises. It is expected that the majority of reserves, however, would continue to be managed by another arm of the new agency, of which SAFE would be a part. There has been no definitive confirmation of whether the new agency will, in fact, be organised along these lines, or how assets might be allocated among the agency's various branches; but, in keeping with the tendency to change policy slowly, Fitch would not expect an immediate large-scale diversification of existing reserve assets.

Where Are the Reserves?

The US Department of the Treasury publishes monthly data on foreign holdings (by country) of US Treasury bills and bonds. Annual figures are provided on foreign holdings (by country) of US Agency bonds, for which interim monthly stocks can be estimated from published data on monthly net purchases.

Fitch estimates that at end-2006, China held US$350bn of US Treasury securities and US$230bn of US Agency debt, accounting for 54% of official foreign-exchange reserves (see chart). These estimates are based on two assumptions. First, Fitch assumes that all Chinese holdings of US Treasury and Agency securities are part of official foreign-exchange reserves, meaning none of the US$182bn in debt security assets in China's IIP at end-2006 is invested in US government debt. Second, it is assumed that none of the estimated US$81bn in long-term US corporate bonds held by Chinese entities is part of official foreign-exchange reserves.

The chart above suggests China's official reserves are becoming more diversified. The share of reserves accounted for by US Treasury and Agency securities peaked at 69% in August 2004, 15% higher than at end-2006. Even so, China remains a large net purchaser of US Treasury debt, and average monthly net purchases were nearly identical in 2005 and 2006, at US$3.1bn. Average net monthly purchases of US Agency securities increased from US$1.9bn in 2005 to US$3.0bn in 2006. The upshot is China continues to purchase an increasing amount of US Treasury and Agency debt, but Chinese reserves are growing more quickly, resulting in a diversification of the stock of reserve holdings.

The Reserve Outlook

The fact that Chinese authorities have agreed to establish an entity to manage a portion of reserves more actively suggests they recognise that current reserve levels are more than adequate for any precautionary purposes. This view can be corroborated by analysing official reserves using typical ratios to measure reserve cover. Three common ratios to assess reserve adequacy are reserves/broad money, reserves/months of imports and reserves/short-term external debt. China's reserves were equivalent to 870% of short-term debt at end-2006, compared to the median for A rated sovereigns of 120%. In terms of import cover, which Fitch broadens to include all external payments in the current account, reserves in China are equivalent to 14.4 months, versus 3.5 months for the A median. China's reserves relative to broad money (M2) were 24% at end-2006, lower than the peer group median of 30%. While the reserves/M2 ratio is below the peer group median, this is arguably the least relevant ratio for China. It is intended to capture the extent to which reserves could cover massive capital flight. Restricted capital outflows already effectively mitigate this risk.

In the context of continued balance of payments strength and only gradual changes in exchange rate policy, Fitch believes the Chinese authorities have been motivated by two considerations in altering their approach to international financial flows - earning greater returns on the large pool of external assets and easing monetary policy challenges associated with the reserve build.

Greater Returns

There is clearly a desire to earn greater returns on official reserves and to diversify reserve assets, at least to some extent. Instead of only accumulating financial assets, China's new investment entity will also likely invest in real assets. Based on gross external assets at end-2005 as reported in the IIP and Fitch's 2006 estimate of income earnings in the current account (investment income only), Chinese assets generated a return of only about 3.6% last year. If the return on investment were 0.5% higher, at 4.1%, that would represent an additional US$6.1bn in investment income, equivalent to about 8% of FDI inflows. The potential income stream from China's external asset stock provides a strong incentive to seek enhanced returns.

Easing Monetary Policy Challenges

Controlling money supply and credit growth have become leading macroeconomic policy challenges. These challenges relate to risks surrounding the misallocation of credit by the banking system - notwithstanding improvements in the banks in recent years - and excess investment spending, both of which could lead to a return of non-performing loans when economic growth slows.

The PBOC has certainly been active in reducing domestic liquidity created by its accumulation of foreign exchange. In 2006, for example, PBOC net foreign-exchange assets increased by CNY2.2 trillion, or 35%. On the liability side of the PBOC balance sheet, reserve money was up by CNY1.3 trillion (21%) and the stock of outstanding PBOC bills increased by CNY944bn (47%). In December 2006, PBOC bills were equivalent to 38% of reserve money, compared to 32% a year earlier and less than 20% at end-2004. The accumulation of foreign reserves and issuance of PBOC bills to manage liquidity could continue for some time, as the interest rate on one-year central bank bills was about 2.8% by the end of 2006, well below PBOC earnings on its US Treasury holdings.

In Fitch's view, however, there are risks to the PBOC balance sheet growing ever larger, particularly with respect to the currency mismatch between assets and liabilities. In addition, the net interest income of the PBOC (taking account only of interest payments on PBOC bills and interest income on US Treasuries) may not always be in the PBOC's favour. By end-2006, the stock of PBOC bills outstanding was larger than China's US Treasury holdings.

While a new reserve management agency can help to increase the return on external assets, it cannot alleviate the monetary policy challenges faced by the PBOC with the continued accumulation of foreign reserves. As long as balance of payments inflows remain strong, capital account outflows are carefully managed and exchange rate appreciation is measured, the PBOC will continue to build its foreign-reserve holdings.

Some PBOC assets will be transformed from direct foreign-exchange holdings to claims on the new investment agency, but the monetary effects of the addition to reserves would have already taken place. That leaves two other possible policy responses to deal with the money supply challenge - there can be greater capital outflows or greater renminbi appreciation. Capital outflows are not new, as evident by the US$487bn stock of non-reserve external assets, but there are initiatives to raise outflows further. In 2005 FDI outflows were US$11.3bn, and Fitch estimates they increased to about US$15bn in 2006. These are still relatively small numbers, but are considerably higher than they were in previous years. Between 1990 and 2004, annual outward FDI never exceeded US$6.8bn, and averaged only US$2.3bn. Chinese policymakers have explicitly supported the overseas expansion of domestic firms, and higher outward FDI can be expected in the years ahead. This trend is also consistent with what Fitch believes to be a desire of the authorities to acquire real external assets in addition to financial assets.

Capital outflows may also pick up modestly under the Qualified Domestic Institutional Investor (QDII) programme launched in 2006. QDII allocated quotas totaling about US$10bn to various financial institutions in China, allowing them to accept funds from residents for investment in fixed-income markets abroad. Again, the amounts involved are modest - and it may not be appealing to Chinese investors if they believe the renminbi will continue to appreciate - but the programme does signal a desire to open up the capital account to outflows.

Exchange Rate Considerations

If the renminbi were allowed to appreciate more quickly, it could help to ease the PBOC's monetary policy challenges by slowing the accumulation of foreign-exchange reserves. This is not a certainty, however, since the prospects of appreciation may attract more capital inflows, requiring the PBOC to buy foreign exchange to prevent an even greater appreciation if the exchange rate were not fully flexible. At the same time, it is not clear whether a faster appreciation would reduce the trade surplus meaningfully, as labour productivity gains appear to be outpacing wage growth in the manufacturing sector, leading to declining unit labour costs and improving international competitiveness. In any event, to the extent that monetary management is a consideration, the Chinese authorities would almost certainly prefer a solution to the accumulation of reserves that involves more capital outflows rather than less current account inflows.

While many international policymakers suggest a faster appreciation is in China's interest to allow greater flexibility in monetary and exchange rate policy, Fitch considers it likely that a more liberalised exchange rate regime will continue to develop only gradually, and the renminbi will appreciate by 5% against the US dollar this year (on an average annual basis). The trade sector is contributing to strong GDP growth, which is a much higher priority than the adoption of a macroeconomic policy framework consistent with those of advanced economies. Under current policy arrangements, China is also accumulating sizeable external assets. This complicates monetary policy to some degree, but not likely to an extent that would motivate a change in exchange rate policy. Finally, with the establishment of a new agency to manage a portion of official external assets, there may be a greater institutional bias towards continued asset accumulation - a bias that would not favour a stronger currency.

Copyright © ChinaForum 2007