8. Develop a global partnership for development
6.5 Overview of private sector fl ows
The UN (2005: 73) opens its discussion of private capital flows by restating the theoretical benefits that should accrue to developing countries as follows:
Standard economic theory argues that international private capital flows will make a major contribution to development to the extent that they will flow from capital-abundant industrialized countries to capital-scarce developing countries, and help to smooth spending throughout the busi- ness cycle in capital-recipient countries.
However, the report goes on to argue that, in recent years, the reality of private flows to developing countries has contradicted this perspective in two ways. First, capital has not moved from developed to developing coun- tries, but instead has tended to flow the other way.17 Second, the volatility and reversibility of capital flows has itself contributed to financial crises in
developing countries, and has not served to ‘smooth’ domestic spending, but has actually made it more volatile. Furthermore, private flows tend to be concentrated in middle-income developing countries, so that the poorest countries may get little or none of the predicted benefits in any event.
Before exploring these issues further, we shall examine the behaviour of private flows, both to the developing world in total and to each of the developing regions individually.
As can be seen from Figure 6.10 above, the total level of net private capital flows to developing countries has shown considerable volatility since the early 1990s (though this certainly does not mean that the situation was one of stability before that point). By 2003, total net flows were a little over US$175 billion, which is about 15% higher than the yearly average for the period 1993–1997 (the first bar on the figure). To put this in context, Figure 6.11 below compares the level of net private flows over this period, with the corresponding level of GDP in developed countries.
In 2006, the figure fell to a little over 0.8% after having risen year-on-year since the turn of the century. However, this should be contrasted with the figures for 1999 and 2000, which at 0.2% or less of developed market GDP was substantially below that envisaged by the UN as necessary for develop- ment purposes. Again, however, this should be contrasted with the 1993–1997 period, when at more than 0.6%, the level of net flows to developing countries was more than double this figure.
What is the regional picture, however?
Figure 6.12 highlights the extremely uneven geographical distribution of private capital flows. Perhaps most striking is the dominant role played by developing Asia and Central and Eastern Europe in this respect: by 2004, these two regions accounted for more than 85% of all capital flows to devel- oping countries. Interestingly, the 1993–1997 period saw relative parity
Figure 6.10 Net private capital flows to developing countries, 1993–2006.
Sources: UN (2005) for 1993–97 average; IMF WEO for 1999–2006.
between developing Asia and Latin America in terms of the total level of private inflows. Since that time, the total flows to Latin America have declined steadily, so that in 2006 they were just a sixth of the average level achieved from 1993 to 1997 and less than 5% of total private capital flows to developing countries.
For Asia, the impact of the Asian crisis can be seen in 1999 and 2000, with net flows being negative. After 2001, however, the recovery was dramatic, so that by 2004 the region received 60% more net private inflows than was the case immediately before the onset of the Asian crisis in 1997.
Figure 6.11 Net private capital flows as percentage of developed countries’ GDP, 1993–2006.
Source: IMF WEO for 2007.
Figure 6.12 Net private capital flows to developing regions, 1993–2006.
Sources: UN (2005) for 1993–97 average; IMF WEO for 1999–2006.
The subsequent decline in flows to the region in 2005 and 2006 is explained in Figure 6.13 above, which disaggregates the flows to emerging Asia, separat- ing out those going to China and India and those going to the rest of the region. As we can see, combined flows to these two economies dwarf those going elsewhere in Asia from 1999 onwards, and the decline in total Asian private inflows from 2005 is entirely the result of the decline in flows to China and India, which more than offset the modest increase in private capital flows to the rest of developing Asia.
What about the composition of inflows, however?
As is shown clearly in Figure 6.14, the volatility of total private sector flows is almost entirely a result of the volatility of (a) portfolio flows, and (b) bank lending. In contrast, FDI flows have shown remarkable stability over the period, even holding up well in the turbulent 1998–2002 period which con- tained the Asian and Russian crises.
In general terms, the composition of capital inflows for all developed coun- tries is replicated in each developing region. In particular, the stability of FDI flows to each region, relative to other forms of capital inflow, is clearly observable. There are some interesting distinctions, though.
Figure 6.15 below illustrates the cumulative inflow for each type of capital for each region, from 1993 to 2004. In each case, the largest form of flow was FDI. However, both in terms of the absolute scale of FDI, and the relative importance of FDI over other types of inflow, the Asian and Latin American regions have clearly been the most important destinations for FDI in the developing world – though, as illustrated in Figure 6.12, the transition econ- omies of Central and Eastern Europe have seen a sharp increase in capital inflows since 2004, and much of that has been in the form of FDI.
Figure 6.13 Net private flows to Asia (ex China and India) vs. flows to China and India, 1999–2006.
Source: IMF WEO.
From 1993 to 2005, FDI flows to all developing regions totalled more than US$1,500 billion. In contrast, bank flows were strongly negative at −US$450 billion, while portfolio flows were positive at a little over US$250 billion.
More than half the negative figure for bank lending was accounted for by the Asian region, reflecting the aftermath of the Asian financial crisis, though bank lending to Africa and Latin America was also negative over the period.
As we have seen, the increase in private capital flows has continued after 2004, reaching US$281 billion by 2006.
One interesting aspect of international financial flows in recent years Figure 6.14 Composition of private capital flows to developing countries, 1993–2004.
Sources: UN (2005) for 1993–97 average; IMF WEO for 1999–2006.
Figure 6.15 Cumulative capital inflows to developing regions, 1993–2004.
Sources: UN, 2005.
relates to changes in the level and distribution of foreign exchange reserves.
Over the past decade current account surpluses in some developing regions – most notably developing Asia, but also Japan – have grown enormously. This has led to a rapid build-up of foreign exchange reserves in these economies, which in turn have been invested in US Treasury bills.
Between 1999 and 2006, an astonishing US$2.75 trillion of foreign exchange reserves were accumulated by developing countries. Figure 6.16 below illustrates the regional distribution of this build up of reserves.
As we can see, the overwhelming majority (67% of the cumulative total) of these reserves have been accumulated in the Asian region, although the Middle East region has also seen a substantial pick-up in reserve accumula- tion since 2002, reflecting the impact of rising oil revenues as the price of oil has risen. Recent years have seen considerable disagreements over the causes of this phenomenon, which has crystallised into a dispute between China and the United States. Box 6.4 describes the nature of this dispute.
Box 6.4 US and China’s exchange rate dispute
More than a decade ago the Chinese authorities pegged the remninbi to the US dollar at a fixed rate. Since then China has experienced annual double-digit growth, driven in large part by a surging export sector, a high proportion of which goes to the US market.
By 2005 the United States’ trade deficit with China had risen to around $200 billion. As we can see from the figure above, the US’s overall trade deficit was almost 6.5% of GDP in the same year. That Figure 6.16 Regional reserve accumulation, 1999–2006.
Sources: IMF WEO.
equates to around $780 billion, so that the deficit with China accounts for around one quarter of the entire US trade deficit.
Despite the fact that China alone cannot account for the scale of the US deficit, many US-based commentators have accused China of
‘mercantilism’ or using an undervalued currency to unfairly increase its share of global markets at the expense of other nations. The testimony of Professor Navarro of the University of California to the Subcom- mittee on Trade of the House Committee on Ways and Means in early 2007 is typical of this position:
By practicing a highly evolved form of 18th century ‘beggar thy neighbor’ mercantilism, China is emerging as a 21st century eco- nomic superpower. While consumers around the world have bene- fited from the flood of cheap goods, China’s broad portfolio of unfair trade practices has resulted in the loss of millions of jobs in countries ranging from the United States and Mexico to Brazil and Lesotho . . . To maintain its undervalued currency – and thereby sell its exports cheap and keep foreign imports dear – China main- tains a fixed currency peg between the U.S. dollar and the yuan. To maintain that peg, China must recycle large sums of its surplus U.S.
dollars gained in the export trade back into the U.S. bond market.
Through such activity, China has become the de facto ‘central banker’ of the U.S. with its net capital inflows roughly equal to that needed to finance the U.S. budget deficit.
The second point here is very interesting. While many have called on China to allow its currency to float – and have proposed various means Current account balances in China and the United States, 1990–2005.
Source: World Bank WDI.
of applying pressure, ranging from the WTO to bilateral trade sanctions to achieve this – if this were to result in China having much lower trade surpluses, then its tendency to invest these surpluses in US Treasury bills would no longer be available to fund the US deficit. What we appear to have, therefore, is a form of ‘Faustian pact’, where both China and the US benefit from the status quo, regardless of whether it is sustainable in the longer term.
As the US dollar continues to slide on global markets in 2008, it is likely that the huge US trade deficit will start to reduce. Furthermore, as China moves to a more flexible peg against a basket of its major trading partners it is likely to be less of a target of US ire. However, China’s move to peg to a basket of currencies may simply be a reflection of its desire to remain competitive in all its major markets, rather than any conversion to the principle of free floating exchange rates determined by the market.
The UN (2005) includes these figures in its total net financial transfer estimates, the result of which is that they describe a huge transfer of private capital from developing to developed countries. Clearly, however, reserve build-up and transfer to US Treasury holdings by (largely Asian) economies is different to other types of capital flow.
The rationale for this course of action is twofold: first, as described in Box 6.4, investing in US financial assets prevents the US dollar from falling further than it otherwise would, which safeguards the competitive export position of East Asian economies, many of whom still attempt to peg their currencies to the dollar. Secondly, countries may be wary of a speculative attack on their currency precipitating a financial crisis. The massive build-up of reserves can therefore be seen as ‘insurance’ against this event – the exist- ence of huge foreign exchange reserves may be sufficient to prevent such an attack, as speculators can see that the exchange rates can be defended with the reserves for a lengthy period.
That said, the situation of developing countries building up large foreign exchange reserves and depositing them in the US does indeed represent a financial transfer on an enormous scale. In effect, the countries of East Asia are lending the US money – at very low rates of interest – to enable the US to purchase their exports. In recent years, economists have developed a rule of thumb for the optimal level of foreign exchange reserves. The ‘Greenspan- Guidotti-Fischer rule’ holds that reserves should be sufficient to cover three months’ worth of imports: today many developing countries – particularly in East Asia – have reserves far in excess of this level, suggesting strongly that the motivation for this build-up relates more to the strategic objectives described above than to any search for optimality.
While this may be in both parties’ interests at the present time, it is a highly precarious, and ultimately unsustainable, global financial situation.
6.5.1 Private capital flows and economic growth
We have seen how there has been something of a boom in private capital flows to emerging markets in the past few years. However, there have been many such booms before, and we must assume there will be many more to come in the future.
An unfortunate fact in this regard is that booms are almost always followed by busts. This cyclicality of private flows to developing countries is particu- larly notable with portfolio and bank flows, which as we have seen, display far more volatility than does FDI. From a theoretical perspective, one would expect the transfer of global capital to developing countries (and therefore to its most productive global use) to result in higher growth in recipient coun- tries, and thus higher global growth. For example, the ‘debt-cycle’ hypothesis states that external inflows raise total savings, which allows an increase in the rate of domestic investment and thus, ultimately, spurs faster economic growth. This in turn will allow the foreign debt accrued to be paid off com- fortably from the proceeds of growth. While this is the theory, Ffrench-Davis and Reisen (1998) list five requirements for this to occur in practice:
1 External flows should be used for investment, rather than consumption purposes.
2 The investment should be efficient.
3 The inflows should be invested in tradable sectors so that a trade surplus is generated, enabling the foreign currency debt to be paid off.
4 Domestic savings should be aggressively mobilised.
5 The ‘virtuous circle’ requires those exporting capital (e.g. from developed countries) to do so in a stable and predicable manner.
Interestingly, these features are more likely to pertain to some types of capital flow rather than others. In particular, long-term FDI is inherently used for investment purposes, whereas portfolio flows in particular may result in appreciating exchange rates and asset prices bubbles, which are likely to raise consumption rather than investment. The same is also true of short-term bank lending used to support and perpetuate asset prices bubbles, as occurred in Thailand in the mid-1990s.
A crucial distinction between these flows is their stability, however. As we have seen, FDI flows have been remarkably consistent since the early 1990s, in sharp contrast to the volatility of other types of inflow.
The UN (2005) describes this volatility in the context of financial market uncertainty as follows:
The basic reason for existence of these patterns is that finance deals with future information that, by its very nature, is not known in advance;
therefore, opinions and expectations about the future rather than factual information dominate financial market decisions. This is compounded by
asymmetries of information that characterize financial markets. Owing to the non-existence or the large asymmetries of information, financial agents rely to a large extent on the ‘information’ provided by the actions of other market agents, leading to interdependence in their behaviour, that is to say, contagion and herding. At the macroeconomic level, the contagion of opinions and expectations about future macroeconomic conditions tends to generate alternating phases of euphoria and panic.
At a microeconomic level, it can result in either permanent or cyclical rationing of lending to market agents that are perceived by the market as risky borrowers.
The problems of uncertainty and asymmetric information described here are also likely to be more prevalent in developing than developed economies, which at least partly explains the greater volatility we see in these markets.
Furthermore, some commentators argue that these ‘boom-bust’ cycles have been exacerbated by the common use of certain modern, quantitative risk management techniques: investors receive the same information processed through the same systems, which encourages them to take the same allocation decisions, thus exacerbating tendencies to herd.18
When these ‘surges’ and ‘droughts’ are combined with the relatively small size of many developing economies in comparison with the scale of global capital flows, we can see how macroeconomic stability can be seriously undermined, with negative effects on growth and poverty reduction.
An important element in this regard is the inability of most developing countries to issue sovereign debt in their own currencies – the so-called ‘ori- ginal sin’ problem.19 While recent years have seen some progress in this area, with international investors increasingly willing to hold local currency debt, there remains a long way to go before the problem is resolved: in 2000 just 6%
of local currency debt issued by developing countries was held by inter- national investors; by 2005 this had risen to 12% – a significant increase certainly, but one that leaves 88% of debt held domestically (IMF, 2005a).
This is more of a problem in some developing regions than others. Clearly, the more developed a country is the more likely it is that investors will be willing to purchase sovereign debt in its own currency. Consequently, for the least developed countries, there is generally no option but to issue debt in foreign currencies.
Figure 6.17 below gives a regional picture of the proportion of outstanding debt denominated in US dollars from 1980 to 2005.
Perhaps the most striking aspect of this figure is its consistency: between around 40% and 60% of debt is denominated in dollars for all regions for most of the twenty-five-year period. Furthermore, if there is a discernible trend it is for this proportion to increase, particularly from 1990.
Figure 6.17 only covers US dollar-denominated debt, however. Historically, developing countries have issued debt in a variety of currencies, although the US dollar has been by far the most important. Since the launch of the euro
though, the dollar has had a serious competitor, and euro-denominated debt has become increasingly popular.
Table 6.2 gives a snapshot of the proportion of outstanding debt denomin- ated in both dollars and euros for 2005. It is clear that in some regions – particularly the Middle East and Africa – the euro has grown rapidly in importance. In combination, we can see that the lowest proportion of dollar/
euro-denominated debt is East Asia, but even here the figure is only slightly less than 65% of all outstanding debt. At the other extreme, in Europe and Central Asia almost 90% of debt is denominated in dollars or euros. Fur- thermore, as shown in Figure 6.17, despite the increasing acceptance of local currency-denominated debt, there is little sign of foreign currency debt decreasing in importance.
This accentuates problems of volatility by creating large and potentially damaging currency mismatches. For example, a developing country whose currency experiences a speculative attack, but where much of its external debt Figure 6.17 Proportion of sovereign debt denominated in US dollars, 1980–2005.
Source: World Bank GDF.
Table 6.2 Proportion of external debt denominated in dollars and euros (%)
Region USD Euro Total
East Asia 55.99 8.5 64.49
Europe/C. Asia 55.68 3.45 89.13
M. East & N. Africa 44.34 31.18 75.52
S. Asia 62.44 5.98 68.42
SS Africa 55.49 18.18 73.67
Source: World Bank GDF.