China's capital account liberalisation in perspective
Speech by André Icard, Deputy General Manager of the BIS, on the occasion of the joint BIS/SAFE seminar on 'Capital account liberalisation in China: international perspectives' in Beijing on 12-13 September 2002.
The Asian financial crisis led to a re-thinking of the issues related to capital account liberalisation. Before that crisis, it had been well understood that sound macroeconomic policies were needed to minimise the risks associated with the opening of the capital account. After that crisis, the critical role played by a well-capitalised, well-managed and well-regulated financial system came into sharper focus. The risks of a highly leveraged corporate structure without effective shareholder discipline also became clearer. Policy-makers gained a better appreciation of the fact that successful capital account liberalisation depends on a broad range of effective policies.
This seminar on "Capital account liberalisation in China: international perspectives" is particularly timely as China prepares for the next phase of financial liberalisation following its recent entry into the World Trade Organisation (WTO).
My introduction seeks to provide the broad background and set the stage for the next two days. I shall first briefly describe the lessons drawn from the experience of capital account liberalisation in other countries. Then, I shall sketch some key features of the Chinese economy and contrast them to the standard scenario which is usually assumed when capital account liberalisation is being discussed. In doing so, I shall argue that China's specific initial conditions pose unusual risks and offer unusual opportunities. Finally, I shall highlight this seminar's main features and outline its programme.
Stylised facts and conventional wisdom
The liberalisation of the capital account has frequently been associated with crises in both industrial and emerging market economies. During the late 1980s and early 1990s, attempts to combine exchange rate stability in Europe with a progressive liberalisation of capital accounts triggered a series of foreign exchange crises. My experience at the Banque de France has left me convinced that it would be a mistake to over-emphasise the permissive factor of capital account liberalisation, while underestimating the importance of weaknesses in fiscal policy as well as incompatibilities between desirable monetary policies and actual exchange rate policies. At the same time, I am also convinced that momentum in capital account liberalisation should be adapted to the degree of resilience of the domestic financial sector to external shocks and its ability to deal with large flows of foreign capital.
In Scandinavia, Latin America and East Asia, capital account liberalisation gave rise to capital inflows that were too large for the domestic financial system to absorb safely. As time passed, capital inflows reversed, and the resulting capital outflows revealed an impaired financial system. The main features of this boom-bust cycle are as follows. Owing to faster growth, higher inflation or both, interest rates tend to be higher in the liberalising economy than in international markets. This interest rate differential combined with the new opportunities offered by liberalisation leads to surging capital inflows, mostly in the form of short-term bank claims or portfolio inflows, as if the floodgate were lifted. This influx of foreign capital in turn leads to currency appreciation under a more flexible exchange rate regime. Under a more stable exchange rate, which may give the false impression that there is little risk to foreign-currency borrowing, capital may enter the country even at an accelerated rate. Either way, the recipient economy can experience rapid monetary growth, and even more rapid credit expansion, asset price bubbles, and booming consumption and investment.
The boom will turn to bust, however, once diminished competitiveness, worsening current accounts and accumulating non-performing loans erode investor confidence. The real appreciation of the currency, whether resulting from nominal appreciation or higher inflation, undermines competitiveness and leads to a deterioration of the current account balance. After asset prices peak, credit extended against property and shares can prove hard to service while much business investment may produce little in terms of cash flow. External imbalance and domestic financial fragility signal higher risks and capital outflows ensue. This reversal puts pressure on both the exchange rate and domestic asset markets and faces the authorities with a dilemma between raising interest rates or allowing the exchange rate to depreciate. This dilemma is even greater if the exchange rate has been kept fixed until the boom turned to bust. Currency depreciation can increase financial distress in the corporate sector heavily exposed to foreign currency loans, especially if depreciation follows a period of fixed exchange rates. At the extreme, such sudden capital flow reversals can be associated with the type of currency and banking crises which convulsed Scandinavia, Latin America and East Asia, notwithstanding some important differences between these episodes.
Several policy implications have been drawn from these stylised facts. First, governments need to pursue sound macroeconomic and trade policies to minimise the risks of capital account liberalisation. Second, economies ought to strengthen their own financial systems and supervisory infrastructure before opening up their capital accounts. Third, a corporate sector marked by fragile finances and poor governance may systematically abuse the opportunities provided by capital account liberalisation. Fourth, many argue that a more flexible exchange rate regime is an important precondition for containing the accumulation of balance sheet risks related to currency mismatches and for preventing exchange market tensions from turning into a full-scale financial crisis. Finally, some draw the conclusion that the capital inflows should be managed or channelled with measures like the Chilean reserve requirements against short-term flows. While there is room for debate about particular policies, the broad lesson is that capital account liberalisation is most likely to succeed if it takes place in a sound macroeconomic environment and if it forms part of a coherent programme of economic reform. Shortcomings in other areas can lead to disastrous results from opening the capital account.
Main characteristics of the Chinese economy
A general lesson drawn from previous experiences is that an economy's particular circumstances often determine the risk profile and policy choices related to capital account liberalisation. China's situation differs markedly from that of countries in Latin America and even some of the Asian economies. Therefore, please allow me to spend a few minutes now to offer an outsider's perspective on some of the important features of the Chinese economy and to consider which elements favour liberalisation and which do not. In this section, I will mainly focus on the technical elements of liberalisation and not any more on general preconditions, such as the foreign exchange regime that I have just discussed.
First of all, even though foreign exchange controls are maintained, China is already largely open to international capital flows. One could even say that China has been more open to FDI than some OECD economies, especially as the country has lately encouraged outward FDI in natural resources and technology. Errors and omissions in its balance of payments have shown relatively large outflows until recently, further indicating considerable cross-border activity. There are growing cross-border flows of renminbi banknotes between China and its neighbours, where the renminbi is sometimes treated as a hard currency. Perhaps the most striking indication of China's financial openness is the marked expansion of onshore foreign currency deposits. Such deposits by Chinese households and firms have reached 8% of local-currency deposits, in part because of the government's policy of keeping onshore dollar deposit rates broadly in line with prevailing international market yields. Monthly changes in these deposits exhibit sensitivity to movements in interest-rate spreads and exchange-rate expectations. These aspects of financial openness favour capital account liberalisation in that they point to substantial adjustments that have already occurred.
The question is sometimes posed whether Chinese officials are prepared to permit money to flow out of China. But the rising dollar deposits, joined by falling foreign currency loans extended by Chinese banks, have already produced a large surplus of foreign currency liquidity in China's banks. These dollars, more than the increase in China's official reserves, contributed to an outflow of some $140 billion during 1999-2001 that eventually found a home in US debt markets and the international banking system. Therefore, the question for China's capital account liberalisation is, in practice, no longer whether funds will be allowed to flow out of China in response to market signals. Rather, the question is whether additional channels should be allowed through which such funds could flow and what conditions need to be fulfilled in order to ensure a smooth move in this direction. Let me first describe what are the characteristics of the Chinese economy which would facilitate the transition to more openness and then indicate the domains in which significant reforms should be undertaken as a precondition to financial liberalisation.
The first and probably the more important positive element is that China's economy has grown rapidly over the past five years, with average GDP growth of 8%. International trade has almost doubled. With among the fastest growth in the world and a gross domestic savings rate of 40% of GDP, the flow of savings in China permits the diversification of domestic portfolios to occur and provide a favourable environment for domestic institutions.
Another circumstance that distinguishes China from many other cases in a manner that favours capital account liberalisation is that domestic interest rates do not at present substantially exceed those in international markets. In the early 1990s, China had experienced moderately high inflation. But a tightening of policy then brought inflation down to a low level. In more recent years, prices in China have fallen mildly. Domestic interest rates in China, still basically set by the government, have come down with inflation to low levels on par and even below those prevailing in international markets.
The situation of the balance of payments and the size of international reserves also represent a big strength: the Chinese currency. The renminbi, became convertible for current account transactions in 1996 and has been tightly tied to the US dollar since 1994. Over the years, China has run small current account surpluses and has enjoyed large net capital inflows on average. A fairly stable inflow of foreign direct investment (FDI) has dominated the capital account and represents the principal counterpart of official foreign reserves of $240 billion. As a result, China's claims on the rest of the world probably exceed its debt to the rest of the world, and in any case the liquidity afforded by its official reserves is very comfortable in relation to its short-term debts. Going forward, China's entry into the WTO is thought likely to promote more FDI inflows even though the WTO-related market opening may put some pressure on its current account. It seems safe to say that China's reliance on stable FDI and its international balance sheet favour capital account liberalisation.
Fiscal policy could prove more of a challenge. Government investment in infrastructure financed by debt has been pursued in recent years, but the stated ratio of government debt to GDP remains manageable. However, the country's under-funded pension system remains a concern as does the banking system's capitalisation. These contingent liabilities suggest a significantly higher effective debt burden.
But the sector which in my view deserves most attention as a potential handicap to a safe capital liberalisation process is the financial one. China's financial system is dominated by a banking sector that is large and liquid, but is not strong. The ratio of broad money to GDP has reached 165%, high by any standard. Chinese banks also enjoy strong liquidity: the loan-to-deposit ratio dropped from 86% in 1999 to 78% in 2001, resulting in large excess reserves. However, the same system is generally regarded as weak in solvency terms, since the reported non-performing loan levels are still high. The Chinese government has taken the first steps to re-capitalise in part the largest banks and to resolve some of the distressed debts. One would agree that more has to be done here in order to reduce potential risks arising from China's capital account liberalisation process.
China's financial system also features new but expanding domestic capital markets. Its stock markets have been growing particularly fast, with a total market capitalisation approaching 45% of GDP. However, the market free float is only a small fraction of the total capitalisation, owing to large government-held stakes in the listed state companies. The equity market has been volatile and richly valued, with prices an average of 50 times earnings. So far, the quality of offerings and range of investors have been limited: listings have been mostly restricted to state firms and retail investors predominate. China's bond market has also grown rapidly, exclusively on the basis of bonds issued by the government and state policy banks that are overwhelmingly held by Chinese banks. Thus, while China's immature capital markets may not favour its capital account liberalisation, certain liberalisation measures could help quicken capital market development.
The implications of initial conditions
The general impression drawn from the preceding section on the characteristics of the Chinese economy is that China's initial conditions do not resemble closely the starting point leading to a stylised capital inflow problem. The two main contrasts between China's particular situation and the stylised facts can be summarised as follows:
- First, China is experiencing stable or slightly falling prices, low nominal interest rates and high valuation of domestic assets, whereas the archetypal inflow scenario features relatively high inflation, high local-currency interest rates and lower asset prices.
- Second, China enjoys strong long-term FDI inflows, abundant foreign currency liquidity and an apparently comfortable position of a net international creditor, rather than suffering from heavy reliance on short-term debt inflows, barely adequate foreign exchange liquidity and rising external indebtedness.
In general, an economy's initial conditions shape the risks of its capital account liberalisation. In particular, the scenario of massive capital inflows upon liberalisation seems less likely and less risky in China. It is possible, I suppose, that FDI inflows could prove excessive. Nevertheless, this situation would lead to a re-examination of the policy that requires such investment to be funded abroad. This policy could be eased if the capital inflow proved to be uncomfortably large.
The main challenge for China probably lies more in successfully managing the liberalisation of capital outflows through more diversified channels. As noted above, this would broaden and institutionalise what we have already witnessed in China: Chinese households and firms switching into dollar deposits, prepaying dollar debts and bundling their capital into informal flows into Hong Kong's stock market. Stepping back, this should be expected, given China's particular situation. With or without liberalisation, market forces in China favour portfolio outflows via more diversified channels, as Chinese residents naturally seek better returns.
China's initial conditions, with incipient capital outflows rather than inflows, pose interesting policy questions. With regard to the currency mix of bank deposits, experience elsewhere suggests that greater exchange-rate flexibility and ongoing interest rate deregulation in China could sharpen the already evident shifts in onshore foreign currency deposits and loans, perhaps leading at times to large outflows. While interest rates could be raised to stem such outflows, their coinciding with falling domestic prices might pose a monetary policy dilemma. To overcome this problem, could these shifts into local foreign-currency accounts be addressed with reserve requirements different from those on domestic currency deposits, as has been done elsewhere? Of course, the answer needs to bear in mind reserve-free deposits available just offshore, thus the measure risks proving ineffective unless some form of capital control remains limiting the free conversion of renminbi into foreign currencies by Chinese residents.
With regard to the gradual liberalisation of two-way equity portfolio flows, how should China strike the balance between regulating and attracting foreign portfolio inflows into China? For instance could China consider setting up a qualified foreign institutional investors (QFII) scheme, on the one hand, and channelling domestic portfolio outflows abroad via some qualified domestic institutional investors (QDII) mechanism on the other? If portfolio outflows prove stronger, might some fee, possibly related to length of holding period, be set and varied to affect such flows in a manner similar to what private mutual funds do to affect shareholder behaviour? Obviously, answers to such questions depend not only on the initial conditions but also on the weights given to various policy objectives and the degree of liberalisation to be achieved. As experienced by some countries, a progressive approach is conceivable with the maintenance of controls and procedures aiming at addressing temporary and specific difficulties. However, these measures should be strong to be effective. They should also be strictly focused to remain compatible with the principle of capital liberalisation, a balance which could prove difficult to achieve and maintain over time. I hope that the diverse liberalisation experiences of the economies represented at this seminar will help Chinese policy makers to ask the right questions and to find sensible answers.
Finally, the idea that China's initial economic conditions could help shape its policy choices with respect to liberalising the capital account is consistent with China's general approach to economic reform. To date, "Chinese characteristics" have played a prominent role in this evolutionary process. Consider the current situation in China where many subsidiaries and joint ventures of foreign companies successfully operate in the country but are still basically forbidden to obtain low-cost funding in the local stock markets or the domestic bond markets. Might China place a high policy priority on reforming and strengthening its capital market and permitting foreign companies operating in China to issue domestic-currency stocks and bonds? Since it is compatible with China's initial conditions of a high level of domestic savings, inviting such foreign-invested firms to participate in the local equity and debt markets is likely to work if the structure of these markets is made appropriate. Moreover, this move would support domestic capital market development and economic growth while complementing China's long-term policy of attracting foreign direct investment.
The seminar programme
This seminar brings to China 13 participants from four continents who have already experienced capital account liberalisation. They will use their experience to address both broad strategy and sector-by-sector policies. Participants on the Chinese side comprise largely those in the State Administration of Foreign Exchange (SAFE) who are directly responsible for implementing current and future policy. Our seminar programme also reflects the importance of an economy's initial conditions in its liberalisation of the capital account. Here, I would like to first highlight two special features of our seminar and then briefly outline the seminar programme for the next two days.
First, our choice of case studies reflects the need to pay closer attention to initial economic conditions. Thus the seminar does not focus on Latin America, as does so much of the recent writing on the subject. This programme is intended to allow Chinese policy makers to sample a richer and more balanced mix of experiences with capital account liberalisation. In particular, I wish to emphasise the cases of Japan and Singapore, where initial conditions appear to have resembled China's.
Second, our seminar focuses on practical considerations and actual experience rather than general macroeconomics. It thus covers five topics: the role of initial conditions and the overall strategy, bank flows, non-bank financial and corporate flows, equity portfolio flows and cross-border cash flows. This sectoral approach reflects the perspective of policy makers and practitioners, since the instruments to manage liberalisation will in practice vary by sector and asset.
Against this general background, Session II explores specifically the role played by initial economic conditions in the overall choice of strategy, focusing on the Japanese experience. As mentioned earlier, when Japan started its liberalisation process, it shared some of the features of China's current environment: low inflation, relatively low interest rates and surpluses in the balance of payments.
Session III focuses on capital flows related to banks. These have posed significant risks in a number of countries. This session addresses the relationship between onshore and offshore money markets, liquidity management of the foreign currency book, and prudential controls on lending foreign currency to residents as well as domestic currency to non-residents. On this subject, the seminar is fortunate to have speakers from Singapore, South Korea and Italy.
Session IV examines issues related to capital flows involving the non-bank financial and corporate sectors. They include both domestic firms tapping offshore funding and foreign companies seeking local currency funding onshore, as well as foreign currency investment by households, pensions, mutual funds and insurance companies. The session draws on the experiences of France, South Africa and the Philippines.
Session V discusses the management of equity portfolio flows. From the perspective of Chinese policy makers, interesting and very topical questions include whether and how to allow QFII, QDII and China depository receipts. The issue of how to merge local- and foreign-currency shares is also topical. The session explores the Malaysian, Indian and Taiwanese experiences.
The penultimate session (Session VI) focuses on offshore circulation of local currency. The Chinese authorities attach great importance to this issue. This is evident from People's Bank of China Governor Dai's recent suggestion that banks in Hong Kong might accept deposits of renminbi currency notes. It is very unusual for a currency to become widely used outside its borders before capital account liberalisation. As a result, this session consults the experience of currencies with little or no living memory of capital controls, namely the Deutsche mark and the US and Hong Kong dollars. The final session is designed to bring together some of the more interesting lessons and ideas from the seminar.
In conclusion, I anticipate and sincerely hope that this joint BIS/SAFE seminar on capital account liberalisation will prove useful to all the participants. The Bank for International Settlements, the economies in the region, and the whole world share an interest in the success of China's capital account liberalisation. Thank you all for your attention.