Ill Winds from Europe*

 
Aug 8th 2012, C.P. Chandrasekhar and Jayati Ghosh

The crisis in Europe intensifies. Spain, which though yet to fully implement the €100 billion bailout of its banks, is preparing for a bailout of its government. Spain's government that needs to borrow as much as €385 billion, is facing bond yields at record highs of as much as 7.75 per cent. That is a combined burden that would be impossible to carry, even for a country with public debt to GDP much lower than Italy. So, another country, after Greece, seems set for a vicious circle of new funding accompanied by austerity; slow growth, high unemployment and low revenues; a collapsing fisc; and one more bailout. As a result, fears of the fallout within and outside Europe are growing. Some evidence is already disconcerting. Italy is widely predicted to be the next victim in Europe. The recovery has lost steam in the US. And the so-called emerging market economies in Asia, especially China and India, which buffered the effects of the previous crisis, are also experiencing a deceleration in growth.

It is the fallout outside Europe, especially Asia, which is now under scrutiny. Two central bankers, from institutions as diverse as the Federal Reserve of Boston and the Bank of Japan, have recently warned that Asia is likely to be affected. This would prolong and intensify the incipient crisis, since these countries are seen as having substantially exhausted the fiscal and monetary headroom they had to deal with a downturn.

The most obvious way in which the European crisis is affecting and would continue to affect the region is through a decline in exports. If we take three significant late industrialisers in Asia-China, India and South Korea-we find that in the case of two of them (China and India) the value of their exports of goods to Europe was between 15 and 16 per cent of the total exports of goods from these countries (Chart 1). That share has fallen between 2009 and 2011, reflecting the effects of the crisis in Europe. In India's case, these developments in the trade in goods would have been aggravated by the slowdown in its services exports to Europe, whose share in India's large exports of services has been rising. Even in those Asian countries where exports to Europe are not as important, the impact of the crisis can be significant. For example, South Korea's exports of goods to Europe amounted to just above 10 per cent of its aggregate exports of goods before the crisis, but even that figure had fallen to 7.3 per cent by 2011.

However, the trade route is only one among the ways in which the contagion effects from Europe can reach Asia. A pathway that could be more important is the increased integration of Asian financial markets with European markets in the aftermath of financial liberalisation in these countries and of the surge in cross-border capital flows that the world had experienced for half a decade before the crisis.

Chart 1 >> (Click to Enlarge)
  
One measure of this kind of integration is the outstanding claims held by European banks in Asia. As estimates from the Bank for International Settlements reported in Chart 2 indicate, such exposure has risen in all three countries being considered here and is large at present. From $35.7 billion, $42.3billion and $52.5 billion in the first quarter of 2005 in China, India and South Korea respectively, European bank claims had risen to $261 billion, $150.6 billion and $161.4 billion respectively by the first quarter of 2012. Moreover, in all cases the exposure of institutions located in Europe was in the neighbourhood of 50 per cent of all international claims on these countries. Unlike in the case of goods exports, South Korea's exposure to Europe on this count is also comparably high. And interestingly, China, considered to have a less open banking sector, has a much higher level of exposure than the other two countries.

Chart 2 >> (Click to Enlarge)

The effects of financial liberalisation on borrowing in these countries are visible in the composition of the counterparties to which European banks are exposed (Chart 3). In all three countries, claims on the public sector are the smallest. In India and South Korea, claims on the non-bank private sector dominate, whereas in the case of China, while claims on the non-bank private sector are substantial, those on the banks dominate overwhelmingly. Clearly, in the case of China, banking liberalisation has encouraged borrowing abroad to finance lending to local borrowers.

Chart 3 >> (Click to Enlarge)

Needless to say, the exposure of foreigners in these countries is not restricted to claims in the form of credit provided to local institutions or agents either from abroad or through an institutional presence in the country concerned. There are other forms of exposure, such as through derivative contracts or the provision of guarantees. As Chart 4 indicates, such exposures can be substantial, and can equal or exceed formal claims on a country and its institutions. Surprisingly, the ambiguously defined category ''other potential exposures'' dominates.

This huge exposure of European banks (and other financial institutions) to Asia, points to the other ways in which the crisis in Europe can impact the Asian region. To start with (as is already clearly happening when we examine European figures excluding the UK), the crisis would require European banks to unwind and reduce their exposure in Asia in order to mobilise resources to cover losses or meet commitments at home. The result is a return flow of capital.

Three kinds of effects are likely to ensue. The first could be a weakening of Asian currencies, as already seen in India, which not only generates instability, but also puts pressure on domestic agents, including firms, with foreign exchange commitments to meet. The local currency outlay they would have to make to meet those commitments would increase, putting pressure on their balance sheets and profit and loss accounts.

Chart 4 >> (Click to Enlarge)

The second would be instability in financial markets, inasmuch as these claims directly or indirectly finance stock market activity. Recently, Eric Rosengren, the President of the Federal Reserve Bank of Boston showed (
http://www.bostonfed.org/news/speeches/rosengr
en/2012/070912/index.htm#figure10
) that greater financial integration was increasing the correlation between the stock returns of large European banks and global banks in China and Japan. In his view, ''while Asian banks did not have a high correlation with U.S. and European bank stock returns during 2007 and early 2008, Asian banks are likely to be more impacted now should a significant shock occur in Europe.'' This only goes to show that ''as interconnectedness increases globally, it will be difficult for any one region to insulate itself from financial strains or crises elsewhere in the world.'' The effects can take many forms. They could for example destabilise financial markets in emerging Asia with unexpected consequences. Or they could, through negative wealth effects, impact adversely on spending and demand. As Hirohide Yamaguchi, the Deputy Governor of the Bank of Japan noted recently (http://www.boj.or.jp/en/annou
ncements/press/koen_2012/ko120709a.htm/
, ''jittery market sentiment … will make Asian economic entities' sentiment cautious, thereby containing their spending.''

Finally, an exit of capital from Asian countries can result in a liquidity crunch that can affect domestic lending adversely, reining in the credit-financed private housing investment and consumption that has come to characterise these countries with adverse implications for demand expansion and growth.

Some combination of these effects does play a role in explaining the slowdown of growth in Asia and could aggravate it in the months to come. This calls for a combination of a fiscal stimulus and an easing of credit provided at reasonable rates of interest. However, fiscal conservatism, an accumulated public debt burden and inflation triggered by the rise in global oil and food prices, is holding back governments and central banks, which are in any case under pressure from global finance to curb spending and tighten monetary policy. That could bring the yet-unfolding crisis in Europe to Asia. In which case these countries would be contributors to rather than buffers against the next crisis, if it materialises.

* This article was originally published in the Business Line on 6 August 2012.

 

Print this Page

 

Site optimised for 800 x 600 and above for Internet Explorer 5 and above
© MACROSCAN 2012