I was very pleased to see that the theme of this year's Think Tank is ‘Securing International Capital Flows’. Capital flows have been an important ongoing topic in regional discussions over the past decade as policy-makers have sought to understand the role these flows played in the Asian financial crisis, and how a recurrence of that episode could be avoided. Seminars such as this make an important contribution to our understanding of the issues.

My remarks this morning are intended to provide some background factual material on capital flows which may help your discussions over the next few days.

Gross International Capital Flows

I will start with global capital flows.

Graph 1 shows gross international capital flows, measured by adding up all inflows of capital into every country in the world, and expressing this as a percentage of world GDP.

There are a couple of points to note about the graph:

• First, after a period of relative stability over the 15 years from 1980 to 1995, international capital flows have increased sharply relative to GDP, almost trebling over the past decade.
• Second, we can identify a number of broad cycles which correspond to the cycles in the world economy. International capital flows tend to rise during periods of strong economic activity and fall back during recessionary times. We should not be surprised by this. International capital flows can be a form of risk-taking, and we know that risk-taking increases during good economic times and is cut back during downturns.

In this context, we should note that capital flows have increased particularly sharply in recent years. While some of this no doubt reflects opportunities created by the pick-up in the world economy, another contributing factor is likely to have been the extraordinary low level of global interest rates. This has encouraged an increase in leverage and risk-taking more generally. An example is the ‘carry trade’, where capital flows from low-interest countries to markets where returns are higher.

Some insights into what is driving capital flows can be gained by looking at the component flows. In Graph 2 global flows are disaggregated into four types:

• direct investment;
• portfolio equity flows;
• debt flows; and
• bank and money market transactions.

It can be seen that, while all categories of capital flows have increased over the past decade, the fastest increases have been in portfolio flows (particularly debt) and banking and money market flows. In other words, the biggest rises have been in flows of institutional money.

We should not be surprised that these flows are rising faster than GDP. Most financial variables – e.g. money and credit, stock market capitalisation, debt market capitalisation – rise faster than GDP. At present, the world's equity and debt markets are twice as large, relative to world GDP, as they were a couple of decades ago (Graph 3). Banking sectors also typically grow much faster than GDP in most countries. This financial deepening is a sign of economic development. Rich countries tend to have higher ratios of financial assets and liabilities to GDP than do poorer countries.

With institutional markets growing relative to GDP, even if the share of those markets being traded internationally were constant, international capital flows would rise relative to GDP.

However, as can be seen in Graph 4, the share of these markets that is traded internationally is not constant – it is rising. Over the past decade, international capital flows have risen faster than the capitalisation of debt and equity markets. That is, ‘home’ bias by investors is decreasing. Twenty years ago, global capital flows represented about 4 per cent of debt and equity market capitalisation; now the ratio is closer to 8 per cent. Again, this should not be surprising, as modern communication and the reduction in barriers to international trade and investment have made it easier for investors to undertake cross-border transactions.

Note, however, that this share is still low, and will no doubt increase further over the years ahead.

Before moving to the topic of regional capital flows, there are a couple of further points worth noting.

First, those who worry about instability in capital flows can draw some comfort from the fact that the recent pick-up in international bank lending has been less biased towards short-term lending than was the case in the lead-up to the Asian crisis. The BIS international bank lending statistics show, for example, that over 40 per cent of international bank loans to emerging markets currently have a maturity of more than one year, whereas in the mid 1990s the respective figure was less than 30 per cent (Graph 5).

Second, and much more importantly, the proportion of international borrowing by emerging markets that is in foreign currencies is declining. This represents a major reduction in their financial vulnerability.

In the lead-up to the Asian financial crisis, 90 per cent of the international bank loans to the region were in foreign currency (Graph 6). This meant that the borrowers were taking virtually all the currency risk associated with this lending. It was not surprising, therefore, that when regional exchange rates fell, many corporations and banks experienced financial difficulties. The proportion of foreign currency loans has since declined to less than 60 per cent – a big improvement on the situation a decade ago.

The importance of having foreign liabilities dominated in domestic currency, rather than foreign currency, cannot be over-emphasised. Australia provides a good example of the benefits that come from doing so. Australia, as you know, has quite large foreign liabilities. It also has an exchange rate that can vary substantially, and around the time of the Asian crisis it fell sharply – as much as some Asian currencies. Yet, despite the high foreign liabilities and falling exchange rate, the Australian economy came through that episode relatively unscathed. While there were no doubt many factors that contributed to that good performance, an important one was that Australians by and large do not borrow in foreign currency.

Australia's foreign liabilities are almost all either in Australian dollars or hedged back to Australian dollars (Graph 7). This means that Australian corporations and banks are not exposed to swings in the exchange rate of the Australian dollar, even though they raise large amounts from foreigners.

Let me draw this section of my talk to a close by summarising the main conclusions from the graphs we have looked at:

1. it is clear that global capital flows are rising much faster than global GDP;
2. the capital flows that are rising the fastest are the ones that have traditionally been seen as the most volatile;
3. these trends are likely to continue because they are being driven by the growth of institutional markets and the reduction in technical and regulatory barriers to international transactions; and
4. borrowers seem to be taking steps to reduce the risks associated with offshore funding, by lengthening the term of loans and shifting the currency denomination to their domestic currency.

These developments carry broad implications for the work of central bankers and other policy officials. One particular point I want to highlight is the need for central bankers to be aware of the risks that their banks and corporations are taking in regard to foreign currency exposures, as these can be a major source of financial vulnerability for a country. That is not to say that firms should be discouraged from seeking capital from around the world. Rather, they should be encouraged to undertake any such borrowing in their domestic currency. One way central banks can contribute to this is by ensuring sound macroeconomic policies, since the ability of residents to borrow overseas in their domestic currency is ultimately determined by the willingness of foreigners to hold that currency.

Growing capital account transactions are also shaping how we view the balance of payments. The traditional view has tended to see causation running from the current account to the capital account – that is, a country for some reason runs a current account deficit or surplus, and this gives rise to capital flows to balance the balance of payments. I think there is evidence that causation is increasingly running the other way; capital account developments are driving current account positions as economies adjust to the ebb and flow of capital. This is part of the general trend for financial developments to become more powerful in driving economies. Going forward, our success as policy-makers is going to depend less on being able to understand exports and imports, and more on being able to understand capital flows.

Regional Capital Flows

Let me now turn to regional capital flows. The three issues I want to focus on are:

• How open is the APEC region to capital flows?
• How important are intra-regional flows in APEC?
• Is capital flowing in the right direction?

(a) Openness

We can get some indication of a region's openness to capital flows by looking at the ratio of foreign assets and liabilities to GDP.

The left-hand panel of Graph 8 shows foreign assets as a per cent of GDP for four regions – the euro area, east Asia, APEC and NAFTA. Apart from the euro area, where the ratio is relatively high at 125 per cent, the other regions all have relatively similar ratios at around 80–90 per cent. The right-hand panel shows the individual ratios for some of the larger APEC countries and, again, at least for the higher-income countries, the ratios are relatively uniform.

What is different is the composition of those assets. In Asia, a lower proportion is held by the private sector and a higher proportion is in official hands. In other words, private investors in Asia are less internationally diversified than those in other regions, while the official sector has larger foreign exposures.

A large part of this is due to China, where controls on the outflow of capital mean that the private sector holds only a very small amount of foreign assets. Even though official reserves are large, the ratio of total foreign assets to GDP for China is only about 40 per cent – i.e. half that of most other countries.

That figure for China looks low, but we don’t have to go back too far in history to see similar figures for other countries. For example, the Chinese ratio is similar to that of Australia a decade ago (Graph 9). Admittedly, a much higher proportion of the Chinese figure is made up of official assets, but it is interesting to note that around the time of the float of the Australian dollar in the mid 1980s the ratio for private foreign assets was similar to that of China now. The similarity no doubt owes to the fact that Australia up to then, like China now, restricted the outflow of private capital.

(b) Intra-regional flows

The next question I want to look at is the importance of intra-regional capital flows, because this provides one indication of the strength of economic ties within the region.

It is hard to get cross-country figures for total capital flows, but the International Monetary Fund does publish fairly comprehensive country classifications of portfolio flows. These are shown in Graph 10.

Europe is clearly very integrated in terms of portfolio investment; over 60 per cent of the total portfolio assets of countries in that region are held in other countries in the region.

APEC is well short of this, at around 35 per cent, but it is still a long way ahead of NAFTA and, most particularly, Asia. In fact, the real story of this graph is how low intra-regional portfolio investment is in Asia. Asian investors allocate only 6 per cent of their international portfolio investments to other Asian countries. This is partly due to the very low intra-regional portfolio investment by Japan.

The reasons for the lack of intra-regional investment in Asia are complex and warrant a speech by themselves. Let me make a couple of brief points, however. The first is that the relative lack of development of Asian financial markets, particularly debt markets, has been one factor slowing intra-regional flows of capital. This is well understood and a lot of co-operative work by regional authorities has been devoted to trying to fix this. What is less well-understood is the way in which macroeconomic factors, such as the level of interest rates, exchange rate policies, and the excess of savings over investment, are working to slow intra-regional flows. This is something that could usefully be discussed over the course of the seminar.

One of the factors lifting APEC's intra-regional share of portfolio investment is the large flows between the US and other APEC countries. The US is both a very important source of, and destination for, portfolio investment. Thirty per cent of US portfolio investment is held in APEC countries (Graph 11). For the APEC countries other than the US, 45 per cent of portfolio investment is within the region, but 35 percentage points of that is in the US. The implication of this is that, if we leave aside the US, APEC countries do not invest much in each other.

(c) Is capital flowing in the right direction?

The final topic I want to touch on is whether capital is flowing in the right direction.

Economic logic suggests that capital should flow from those countries where returns on investment and savings are low to those where returns are high. As older, more developed economies would on balance be expected to have lower returns than emerging economies, it follows that the general pattern of global capital flows should be from the mature economies to emerging economies.

I would contend that, by and large, this is the pattern observed in private capital flows. Europe and Japan have outflows of private capital, while non-Japan Asia has inflows (Graph 12). Private capital is, however, also flowing heavily into the US. Various reasons have been put forward for why that is so:

• the US economy is relatively fast-growing by the standards of developed economies, and therefore on balance provides relatively high returns on both debt and equity; and
• US markets offer a combination of size, sophistication, liquidity and investor protection that is not readily available elsewhere.

Official capital on the other hand is flowing heavily out of Asia, more than offsetting the inflows of private capital. This is the result of authorities in the region accumulating reserve holdings in the developed countries, particularly the US (Graph 13). This outflow of official capital from the Asian region is another topic which this Group might want to spend some time considering over the next few days. What are the benefits and costs arising from the authorities when undertaking such activities, and is this outflow from the region sustainable?

Conclusion

The organisers of this year's Think Tank have chosen a very important topic for your discussions. Cross-border capital flows are growing and they are having an increased bearing on our work as policy-makers. We should not try to limit these flows – that would be both pointless and harmful. Rather, we need to understand them and learn to live with them. Among other things, that means following sound policies and promoting the development of financial systems that allow banks and corporations to manage the risks involved in cross-border flows.

I would encourage you to have a robust discussion of these issues over the course of your seminar.