World overview and European sovereign debt.

Holland, Dawn ; Barrell, Ray ; Delannoy, Aurelie 等

Short-tem inflationary pressures have risen on a global scale in recent months and, given the source of the impulse is commodity markets, this dampens the prospects for growth in 2011 in most countries. Global food prices have been under pressure since July 2010, reflecting poor harvests in many parts of the world. Metals prices have also risen rapidly, while non-food agricultural price inflation accelerated towards the end of the year. We expect average food and other agricultural prices in 2011 to be more than 25 per cent higher than they were in 2010, while metals prices are expected to be more than 30 per cent above last year's average level, as shown in figure 1. The price of oil exhibited moderate inflation through September 2010, but rose sharply in the final quarter of the year. The rise in the price of oil may be a reflection of demand pressures from countries such as China and India, as well as the recovery in demand from advanced economies, while the weakness of the US dollar and an expected tightening of regulation following recent oil spills may also be adding to price pressures. The price of Brent crude currently stands at over $98 per barrel, roughly $19 per barrel higher than was priced into futures markets three months ago. Barrell, Delannoy and Holland discuss the macroeconomic implications of the recent rise in the oil price elsewhere in this Review. If sustained we expect this to reduce growth in the OECD by about 1/2 per cent this year. The impact on oil-intensive emerging economies such as China and India may be slightly greater, while oil exporters gain from the high price.

For the most part the rise in commodity prices does not yet appear to have spread into wage setting in the advanced economies, although real wages in France and Japan have risen more rapidly than elsewhere. Employment prospects remain weak in many countries, with the unemployment rate edging up in the final quarter of 2010 in Denmark, France, Italy, Austria, Spain, the UK and the US. This largely offsets the risks of a wage-price spiral developing, but monetary authorities should keep a close watch on wage agreements over the next several months. Wage bargainers in energy importing countries may be aware that a rise in oil prices reduces the real wage they can achieve as it changes the terms of trade against them.

Global inflation is expected to accelerate to 4 1/2 per cent in 2011, from 4 per cent last year, while consumer price inflation in the OECD group of economies is forecast to rise to 2.2 per cent in 2011, compared to our forecast of 2 per cent in October 2010. Heightened price pressures have compounded the partial withdrawal of economic stimulus packages, primarily in Europe and China, and NIESR's estimates suggest that world GDP growth slowed to an annualised rate of 3.7 per cent in the second half of 2010, from 5.7 per cent in the first half of the year. We have revised our forecast for world GDP growth in 2011 down to 4.2 per cent, from a projection of 4.5 per cent three months ago.

[FIGURE 1 OMITTED]

The impact of higher inflation on global growth has been moderated to some extent by the relaxation of both fiscal and monetary policy in the US and Japan towards the end of last year. This is in contrast to the fiscal tightening measures that have been implemented across most of Europe. While monetary policy remains accommodative in most of the advanced economies, the rise in inflationary expectations has pushed nominal long-term interest rates up from the recent historical lows reached in August 2010. Where economic recovery is more entrenched, such as in Australia, Brazil, Canada, China, India, Korea, Norway, Poland and Sweden, central banks have already started to normalise the monetary policy stance and increase interest rates.

The loosened monetary stance in the US, which entails both a new round of quantitative easing and a softer stance on medium-term inflation, is roughly equivalent to a 100 basis point cut in interest rates (see the discussion by Barrell, Delannoy and Holland in this Review), and has widened expected global interest rate differentials. This has put downward pressure on the US dollar, and upward pressure on exchange rates in countries that have begun to reduce their monetary stimulus. The consequent exchange rate adjustments have sparked concerns of a global currency war, and a number of countries, including Japan, Brazil, Korea, Taiwan, and Russia, intervened in currency markets in the latter part of 2010 to stem the rise in their exchange rates. The impact of these interventions tends to be short-lived, but they may smooth the speed of capital inflows. Since the second quarter of 2010, the US dollar has depreciated by more than 10 per cent against the currencies of Australia, Japan, Sweden and Switzerland, while currencies in Brazil, Canada, Korea, the Euro Area, the Czech Republic, Poland, Taiwan, the UK, New Zealand, South Africa and Norway have appreciated by 4-10 per cent against the dollar.

Despite an 8 per cent rise in the yen last year, we estimate that Japanese exports rose by over 25 per cent, allowing GDP to expand by 4 1/2 per cent in 2010, one of the fastest rates of growth in the OECD. Germany also recorded exceptionally strong growth of about 3 1/2 per cent last year, with both countries benefitting from strong import demand in China. Nonetheless, output in Germany and Japan remained 1.8 and 3.4 per cent, respectively, below pre-crisis peak levels in the third quarter of 2010, as illustrated in figure 2, while export volumes remained 1.4 and 7 per cent below pre-crisis peak levels. Of the G7 economies, only those in North America had regained pre-crisis levels of output by the end of 2010. Investment remains the biggest drag in most countries, with the level of private sector investment some 5 to 25 per cent below pre-crisis levels. In recent months bank lending has edged up and investment is expected to support growth in most countries this year.

[FIGURE 2 OMITTED]

Consumer spending in Canada and France exceeded pre-crisis levels by the third quarter of 2010, partly a reflection of temporary income tax cuts introduced through fiscal stimulus packages in the wake of the financial crisis. Consumer spending in Germany, Japan and the US is also expected to have reached pre-crisis levels by the end of 2010, while consumption in Italy and the UK remains disappointing. Going forward, household spending will be restrained by the removal of temporary tax stimuli in a number of countries, but housing markets show signs of recovery, with some house price rises in the US, France, Finland, Canada, Belgium, Australia and Denmark, which should offset some of the impact of higher taxes. Housing markets in Greece, Ireland, Austria and Spain, however, are still experiencing price falls and/or declines in housing investment.

European Sovereign Debt

Fiscal consolidation in Europe has progressed more rapidly than in North America and Asia. This is acting as a strong restraint on the recovery in a number of European economies, including Greece, Ireland, Portugal, Spain and the UK, where fiscal consolidations worth 4 to 8 per cent of GDP are planned for 2010-12. Early consolidation in the former four reflects the recent difficulties in raising funds on the open market to cover the financing needs of the general government. The yield on 10-year government bonds in Greece currently stands at over 11 per cent, while it stands at 8 3/4 per cent in Ireland, nearly 7 per cent in Portugal and over 5 per cent in Spain, compared to just 3.2 per cent in Germany.

The variance of bond spreads within the Euro Area began to widen at the onset of the financial crisis in 2008, but remained within broadly containable bounds until the spring of 2010. Figure 3 illustrates the standard deviation of 10-year government bond spreads over Germany in the Euro Area (1) since 2006. Prior to the financial crisis, there were only marginal differences in government bond yields across countries, as the common exchange rate was thought to fully cover the risk of holding sovereign debt from any Euro Area country. When the financial crisis hit and it became clear that the banking sectors in some countries were more fragile than others, investors began to differentiate Euro Area sovereign debt. As financial markets stabilised, spreads in the Euro Area narrowed over much of 2009, but Greek bond spreads widened towards the end of the year as new estimates revealed that the size of the government deficit in 2008 had been grossly underestimated. Meanwhile bond spreads in Ireland had already widened, reflecting the potential costs of bank bail-outs, which left the government with over 150 per cent of GDP in contingent liabilities.

[FIGURE 3 OMITTED]

The Greek crisis deepened as the initial fiscal consolidation programme put forward lacked credibility, compounded by strikes held in protest of the announced austerity measures. Following a brief respite in May 2010, backed by support in the form of guarantees from the ECB and the IMF, Greek spreads continued to widen throughout most of the second half of the year, despite an improvement in public finances of an estimated 6 per cent of GDP in 2010. By the time of the bailout of the Irish banking system in November 2010, the sovereign debt crisis had spread to Portugal and Spain, with some rise in bond spreads in Italy and Belgium also becoming apparent. The standard deviation of bond spreads within the Euro Area reached a peak in early January 2011, and has since receded marginally.

The cases of Greece and Ireland should be viewed as distinct from one another. The primary cause of the loss of confidence in Greek sovereign debt lies in the lack of credibility of the government sector along with a large and sustained balance of payments deficit, whereas the Irish crisis resulted from excessive losses in the banking system. While the situation in Portugal is closer to the Greek model and the Spanish case is closer to the Irish model, it is not clear that either country suffers from the same scale of problems, and the rise in government bond spreads in these economies, as well as in Italy and Belgium, may owe as much to contagion as to market fundamentals.

Figure 4 illustrates the gross financing needs of several governments in 2010 and 2011, to give an indication of the relative magnitude of funds required by those economies where sovereign bonds have come under pressure. The figure distinguishes between funds required to pay back maturing debt and funds required to finance the current fiscal deficit. The first thing to note is that the financing needs of the Euro Area countries are dwarfed by that of Japan, where 10-year government bond yields stand at just over 1 per cent per annum. Financing needs in the Euro Area are also below those required in the US, (2) where bond yields are only marginally higher than those in Germany. Clearly the magnitudes required are not exorbitant within a global context.

[FIGURE 4 OMITTED]

Nonetheless, within the Euro Area, only in Finland is the borrowing requirement in 2010 or 2011 below that of Germany. While Greece is expected to have the highest borrowing requirement in 2011, yield spreads do not otherwise track borrowing needs very closely. If the total borrowing requirement is adjusted by the share of debt held by non-residents, this strengthens the correlation with the yield spreads, indicating that it may be more difficult to convince non-residents to roll-over maturing debt than residents. According to OECD and IMF figures, more than 70 per cent of the central government debt stock in Ireland, Portugal and Greece is held abroad, while less than 60 per cent of government debt in Italy, Belgium, Germany and Spain is held abroad. Yield spreads are also closely correlated with the current account balance. Large current account deficits in Greece, Spain and Portugal suggest that a smaller share of the financing needs of the government can be met through domestic savings. There is also a closer relationship between bond spreads and the expected fiscal deficit in 2011 than the expected total borrowing requirement in 2011. This may reflect the relative ease of financing roll-over debt compared to new issue.

[FIGURE 5 OMITTED]

There are clearly some fundamental factors underlying the Euro Area bond spreads, although a significant margin appears to reflect an unexplained loss of confidence, which demands a high risk premium. The fact that spreads have not receded, despite the introduction of stringent fiscal consolidation programmes, illustrates the difficulty in restoring credibility. The costs of this loss of confidence are high. While our forecast assumes that yield spreads gradually revert to historical norms, this process is expected to be protracted. In our central forecast, we allow spreads in Spain to revert to historical levels by end-2015, those in Portugal to come down by end-2017, while spreads in Ireland and Greece are assumed to remain elevated beyond the end of our ten-year forecast horizon, reflecting the high costs in terms of credibility of requiring an external bail-out.

Figure 5 illustrates the difference between the current 10-year bond yield, which can be thought of as the marginal cost of borrowing, and the average rate of return paid on the debt stock, calculated as total general government interest payments relative to the size of the debt stock. Over much of the past decade the marginal rate has exceeded the average rate in Ireland, France and Portugal, while the average rate has been higher in Greece, the Netherlands, Belgium and Germany. These differences are largely a reflection of the average years to maturity of the debt stock and the relatively high rates on bonds issued in the 1990s and do not necessarily represent the costs of re-issuing maturing debt. However, the sharp rise in marginal interest rates in some economies in 2009-10, at a time when government debt is also rising rapidly, has serious implications for the interest liabilities of governments going forward.

In order to assess the impact of the rise in bond spreads on deficit positions in the Euro Area, we ran a model simulation using NiGEM to calibrate the potential savings that could be made were bond spreads to revert to historical norms in 2011 rather than in 2015 and beyond, as assumed in our baseline forecast. This results in a decline in government interest payments, which are modelled following a perpetual inventory framework. The change in interest payments is determined by the size of the debt stock, the average term to maturity of the debt stock and the difference between the marginal interest rate prevailing at the average historical period of issue of maturing debt and the current marginal rate. Lower interest payments improve the fiscal position, allowing the stock of debt to recede. Figure 6 illustrates the expected change in budget deficits and government debt stocks by 2014 in the sensitive Euro Area economies if bond spreads were to revert to historical norms immediately with no underlying change in inflation expectations. Greece would see interest rates decline by 8 percentage points, allowing the deficit to improve by 10 per cent of GDP by 2014. The debt stock would fall by 6 1/2 per cent of GDP relative to our forecast baseline by 2014, and would continue to improve over the next decade. Budget deficits in Ireland and Portugal would improve by 2 3/4 per cent of GDP by 2014, with marginal improvements in the budget positions in Spain and Italy, where government bond spreads remain relatively low.

[FIGURE 6 OMITTED]

An improvement in the Greek budget balance of 10 per cent of GDP relative to the baseline would leave a budget surplus of close to 5 per cent of GDP from 2014. While it may be wise to run a surplus for several years in order to bring the debt stock to within sustainable bounds (see the discussion on the UK in the Commentary in this Review) there would be a strong argument for using the surplus to alleviate some of the contractionary effects of the austere budget policies introduced last year. If the government were to target a balanced budget rather than a surplus of 5 per cent of GDP, this would allow the income tax rate to come down by 10 cents on the euro and increase GDP growth by an average of 1/2 per cent per annum over the next decade. This is an estimate of the costs faced by the Greek taxpayer due to the loss of credibility of the government sector, which was brought about by repeatedly underestimating the size of the fiscal deficit.

High government bond spreads could prove much more costly if the risk premium on government debt were to spillover into private sector borrowing costs. Based on the wedge between bank lending and deposit rates, there is so far no evidence in Greece, Portugal or Spain of a related rise in the risk premium changed by banks to the private sector. Ireland faces a greater risk of spillovers to private sector borrowing rates, as the banking system is sustained by government guarantees. Our model simulations suggest that a 1 point permanent rise in private sector borrowing costs in Ireland would reduce output by 0.1 per cent in the first year and more than 1 per cent in the long run.

DOI: 10.1177/0027950111401130

Table 1. Forecast summary

Percentage change

 Real GDP (a)

 World OECD China EU-27 Euro USA Japan
 Area

2007 5.3 2.7 13.2 3.0 2.8 1.9 2.3
2008 2.8 0.3 9.3 0.4 0.3 0.0 -1.2
2009 -0.6 -3.4 8.9 -4.2 -4.0 -2.6 -6.3
2010 5.0 2.9 10.2 1.8 1.7 2.9 4.5
2011 4.2 2.4 9.0 I.8 1.7 2.6 2.1
2012 4.2 2.5 8.1 2.1 2.0 2.7 1.4
2001-06 3.9 2.3 9.3 2.0 1.8 2.4 1.4
2013-17 3.9 2.5 7.6 2.1 1.9 2.7 1.7

 Real GDP (a)
 World
 Germany France Italy UK Canada trade (b)

2007 2.8 2.3 1.4 2.7 2.2 7.4
2008 0.7 0.1 -1.3 -0.1 0.5 2.8
2009 -4.7 -2.5 -5.1 -4.9 -2.5 -11.1
2010 3.6 1.5 1.0 1.4 2.9 12.1
2011 2.6 1.7 1.1 1.5 2.7 7.8
2012 2.3 2.0 1.5 1.8 2.7 5.7
2001-06 1.1 1.8 1.1 2.5 2.6 6.7
2013-17 1.8 1.8 1.7 2.4 2.6 5.2

 Private consumption deflator

 OECD Euro USA Japan Germany France Italy
 Area

2007 2.2 2.3 2.7 -0.6 1.8 2.0 2.3
2008 2.9 2.7 3.3 0.4 1.7 2.9 3.2
2009 0.3 -0.2 0.2 -2.1 0.1 -0.6 -0.2
2010 1.7 1.7 1.7 -1.5 1.8 1.3 1.7
2011 2.2 2.1 2.2 -0.1 1.8 2.4 2.3
2012 2.0 1.9 2.1 0.2 1.8 1.6 2.0
2001-06 1.9 2.1 2.3 -0.9 1.4 1.7 2.6
2013-17 2.0 2.0 2.2 1.4 1.8 1.9 2.0

 Private
 consumption
 deflator Interest rates (c) Oil
 ($ per
 UK Canada USA Japan Euro barrel)
 Area (d)

2007 2.9 1.6 5.1 0.5 3.8 70.5
2008 3.1 1.6 2.1 0.5 3.9 95.7
2009 1.3 0.5 0.3 0.1 1.3 61.8
2010 4.2 1.5 0.3 0.1 1.0 78.9
2011 4.2 2.4 0.4 0.1 1.3 97.4
2012 1.8 1.6 0.9 0.2 2.2 106.5
2001-06 2.0 1.7 2.7 0.2 2.8 37.8
2013-17 1.9 2.1 3.0 0.8 4.2 122.9

Notes: Forecast produced using the NiGEM model. (a) GDP growth at
market prices. Regional aggregates are based on PPP shares.
(b) Trade in goods and services. (c) Central bank intervention
rate, period average. (d) Average of Dubai and Brent spot prices.