Investment, Savings and Growth - International Experience in Relation to Some Current Economic Issues Facing China

The following study on the international relation of investment, savings and economic growth is based on a paper produced by the author for Antai College of Economics and Management, Jiao Tong University Shanghai.

* * *

Introduction

This is the first of two papers devoted to the evidence on the relation between investment, savings and growth with particular regard to present economic issues facing China in relation to the international financial crisis. The two papers, although interrelated, are produced separately for the following reasons.

The first paper is an historical and comparative examination of the factual relation between investment rates and economic growth rates. The economic evidence it produces is clear. A very high rate of investment is required for rapid economic growth of the 8% a year level China requires. It is a high level of investment, not a high level of consumption, which is indispensable for rapid economic growth rates – there are no examples of countries with low rates of investment and very high economic growth. Those who argue that China, to maintain its level of economic growth, must increase its consumption level and reduce its rate of investment must produce evidence to justify that claim – and will be unable to do so.

However, from an underlying strategic economic issue such as the above, it is not possible in a one to one mechanical way to derive immediate policy conclusions – something the present author is acutely aware of from both theoretical considerations and practical experience. In order to judge a specific immediate policy it is necessary to have not only an overall framework but also detailed knowledge of concrete economic circumstances. The issues dealt with here affect economic strategy and other concrete factors must be taken into account in framing short term economic policy responses.

Investment and savings

In relation to the international financial crisis significant discussion has taken place regarding the US and China’s savings rate. However, from the point of view of China’s economic growth rate, the issue with the most direct effect is China’s rate of investment. The savings’ rate’s effect on growth is indirect.

This distinction may be easily illustrated by noting that while savings are necessarily required to finance investment it is both theoretically and practically possible, for example, for a country to have a high savings rate but to have relatively low or moderate investment and economic growth rates – Saudi Arabia and Libya are examples. In such cases a high savings rate is not used to maintain a high rate of domestic investment, with an associated high rate of economic growth, but instead foreign assets or exchange reserves are accumulated.

It is therefore investment which directly affects the rate of national economic growth. For that reason, regarding the potential for strategic economic growth, analysis should commence with the investment rate.

Confusion of domestic demand and domestic consumption

This strategic issue relates to a further, more immediate, economic question. In sections of the media stimulation of ‘domestic demand’ is sometimes treated as though it were the same issue as the stimulation of ‘domestic consumption’. This is self-evidently theoretically false. Domestic demand consists of two components, investment and consumption. Stimulation of domestic investment is just as much stimulation of domestic demand as is stimulation of domestic consumption.

The consequences of different allocations of GDP resources to investment and consumption are, however, extremely different from the point of view of China’s economic growth. As will be seen in detail below a very high level of fixed investment in GDP is a precondition for a high economic growth rate in any country - including China. Lowering the proportion of China’s investment in GDP, that is raising the proportion of consumption in GDP, will lead to a much slower rate of growth of China’s GDP. From this more immediate angle also the first key macro-economic issue that should be examined is the investment rate.

This paper, therefore, examines the relation of the rate of investment and the rate of economic growth both from a fundamental historical perspective and from the point of view of the recent international experience of high growth rate economies.

The tendency of the proportion of the economy devoted to fixed investment to rise

Considering first the investment rate from a long term historical perspective, one of the most factually well established historical trends of economics is that the proportion of the economy devoted to fixed investment historically rises with time - and that this rise is correlated with increasingly rapid rates of economic growth.

This process can be clearly measured over a three hundred year period, and can also be seen to operate dramatically in the period since World War II. All major economies that have grown rapidly have a high level of fixed investment. There are no examples of major economies which have grown rapidly with a low rate of investment.

These facts have evident conclusions for the discussion of the model of economic growth and for China’s investment level.

After considering this from the point of view of a long timescale, setting out the factual data, it will be examined from the point of view of the experience of high growth economies since World War II.

The historical trend of the proportion of investment in GDP

Figure 1 shows the percentage of fixed investment (gross fixed capital formation) in GDP for a series of major countries over the longest periods of time for which data is available.[1]

Figure 1

The pattern is evidently clear and striking. By far the strongest trend is for the proportion of GDP devoted to fixed investment (gross domestic fixed capital formation) to rise with time. This in turn, as will be shown, is associated with progressively rising rates of economic growth.

The historical correlation of increasing proportions of GDP devoted to investment with rising rates of GDP growth

Considering this historical trend in more detail, and analysing countries in the chronological order in which a new peak in the proportion of GDP devoted to gross fixed domestic capital formation appeared, the following is the historical pattern.

- Commencing with the period immediately antedating the industrial revolution, the proportion of GDP devoted to fixed investment in England and Wales, at the end of the 17th century, was 5-7 per cent. [2] This rose slightly, although current estimates are that it did not rise greatly, during the 19th century - peaking at over ten percent of UK GDP prior to World War I.

This level of investment was sufficient to launch the first industrialisation of any country but at a rate of growth which, while unprecedented at the time, was extremely slow by contemporary international standards - about two per cent a year.

- Turning to the latter part of the 19th century, the proportion of US GDP devoted to fixed investment had risen to considerably exceed that for the UK – reaching a level of 18-20 per cent of GDP by the last decades of the century.

A sharp fall in the proportion of the US economy devoted to fixed investment commenced in the late 19th century, and was particularly pronounced during the period between World War I and World War II – being associated with the great depression of the inter-war period. After World War II the US resumed its pattern of 18-20 per cent of GDP being devoted to gross fixed capital formation. This generated an average growth rate of 3.5 per cent a year. With such a growth rate an economy doubles in size every 20 years and quadruples in size every 40 years. It was on the basis of this historical level of investment, and growth rate, that the US overtook Britain to become the world’s greatest economic power.

- In the period following World War II Germany achieved a level of fixed investment exceeding 25 per cent of GDP – peaking at 26.6 per cent in 1964. This period 1951-64 was that of the post-war German ‘economic miracle’ with average growth of 6.8 per cent a year - with such a growth rate an economy doubles in size every 11 years and quadruples in 22 years.

- Starting at the beginning of the 1960s Japan achieved a level of gross domestic fixed capital formation of more than 30 per cent of GDP. This reached a peak in the early 1970s, at 35 per cent of GDP, before later sharply falling. During the period of a high and rising rate of investment in GDP the average annual rate of growth of the Japanese economy was 8.6 per cent.

- From the 1970s onwards, South Korea similarly achieved a level of fixed investment of 30 per cent of GDP. During the 1980s this rose above 35 per cent of GDP. The other East Asian ‘Tiger’ economies – Singapore, Hong Kong and Taiwan – showed a similar pattern. South Korea’s economy confirmed the relation between fixed investment and economic growth illustrated by Japan by growing in this period by an average 8.3 per cent a year.

Such growth rates in Asia showed that something unprecedented in human history was now possible – that it was possible to industrialise an economy, and achieve a ‘first world’ level of development, in a single generation.

- From the early 1990s onwards China achieved sustained rates of fixed investment of 35 per cent of GDP with, from the beginning of the 21st century, this rising to more than 40 per cent of GDP – a level never before witnessed in human history. The result was average 9.8 per cent a year economic growth over a sustained period – also the most rapid sustained economic growth ever seen in human history.

- To complete the chronological picture, the proportion of GDP devoted to fixed investment for two countries recently undergoing rapid economic growth, India and Vietnam, is shown. The proportion of Indian GDP devoted to fixed investment has not reached the Chinese level but has become high – reaching 33.7% of GDP in 2007 and 37.6% of GDP by the second quarter of 2008. On this basis, in the last five years, India has achieved an average growth rate of 8.8 per cent a year.

In Vietnam the proportion of GDP devoted to fixed investment rose from 13 per cent in 1990 to 25 per cent in 1995 and then to 37 per cent in 2007. Economic growth has accelerated rapidly, rising to an average of 7.9 per cent a year in the five years up to 2007.

Considering these trends, such a high level of investment is a necessary condition for rapid economic growth. No substantial country without comparable high levels of fixed investment has achieved such rapid rates of growth on a sustained basis. But while such a high level of investment is a necessary condition for rapid economic growth it is not a sufficient condition. Other elements which must accompany a very high rate of investment in GDP to produce rapid economic growth are considered below.

Recent experience of countries with high rates of economic growth

Turning to analysing post-World War II examples of sustained high economic growth, only 21 countries have achieved 8% growth a year sustained over a 20 year period since World War II. Leaving aside two extremely small states, Botswana (population 1.6 million) and Swaziland (population 1.1 million), which have economies dominated by individual projects, these countries that have undergone at least an 8% growth rate over a twenty year period fall into only two categories.

The first are eight Asian economies which have experienced prolonged periods of rapid growth - China, Japan, Singapore, South Korea, North Korea, Taiwan, Thailand, and Hong Kong. These form the primary focus of this study as their economies are not dominated by direct and indirect effects of the single commodity oil.

The second group are oil producers, or states adjacent to oil producers, in which rapid economic growth has been due to the direct and indirect effects of producing this commodity.[3] Growth rates based on oil are evidently not available to countries that do not have oil reserves and therefore do not form a generalisable model of development or are not immediately adjacent to countries which are large oil producers – for this reason the growth pattern of economies dominated by oil production are not considered in detail here.

Investment levels in the high growth Asian economies

To illustrate the decisive role played by high investment rates in sustaining high growth rates the percentage of Gross Fixed Capital Formation (fixed investment) in GDP for six of the eight high growth Asian economies is shown in Figure 2 - comparable IMF data for North Korea and Taiwan is not available. India has been added to this comparison due to the size of its economy and its recent rapid growth.

The evident feature of these economies countries is that all have had, during their periods of rapid growth, very high percentages of Gross Domestic Fixed Capital Formation in GDP. Taking the peak years for each country, Gross Domestic Fixed Capital Formation reached 35.6% of GDP in Hong Kong, 36.4% of GDP in Japan, 39.1% of GDP in South Korea, 41.6% of GDP in Thailand, 42.7% of GDP in China and 47.4% of GDP in Singapore.

Figure 2

It may be seen that no cases at all of rapid sustained economic growth without such a high rate of investment are to be found in such high growth economies. It is therefore evident that the economic evidence demonstrates that a high percentage of gross domestic fixed capital formation is a precondition for rapid economic growth. It is a high proportion of investment in GDP, not a high proportion of consumption, that forms the precondition for rapid economic growth.

Furthermore detailed examination makes clear that in those countries in which investment declined as a proportion of GDP – Japan, Hong Kong, South Korea and Singapore – this led to a marked decline in economic growth. On the contrary in those economies in which investment rose as a percentage of GDP, China and India, economic growth accelerated. The correlation between a high rate of growth and a high rate of investment is therefore evident.

In the data below the annual rate of growth is stated as the average over a five year period - in order to smooth out purely short term fluctuations in business cycles.

Japan

Measured in PPP terms Japan is Asia’s second largest economy. Japan’s rate of Gross Domestic Fixed Capital Formation peaked at 36.4% of GDP in 1973 but then fell to 23.3% of GDP by 2007.

Over the same period of time Japan’s annual rate of GDP declined from the 8.4% per cent rate in 1973 to only 2.1% (see Figures 3 and 4).

Figure 3

Figure 4

South Korea

South Korea is the Asia’s 4th largest economy - after China, Japan and India. South Korea’s level of Gross Domestic Fixed Capital Formation in GDP peaked at 39.1% in 1991, although the 1996 level of 37.5% was only marginally lower.

Thereafter South Korea’s level of Gross Fixed Capital Formation declined sharply to 28.8% of GDP in 2007. South Korea’s annual rate of GDP growth fell in parallel from 9.4% in 1991, and 7.3% in 1996, to 4.4% in 2007 (see Figures 5 and 6).

Figure 5

Figure 6

Thailand

Thailand’s percentage of Gross Domestic Fixed Capital Formation peaked at 41.6% of GDP in 1990 and 41.1% of GDP in 1995. It then fell to 26.8% of GDP in 2007.

Thailand’s annual rate of GDP growth over the same period fell from 10.9% in 1991, and 8.6% in 1995, to 5.6% in 2007 (see Figures 7 and 8).

Figure 7

Figure 8

Singapore

Singapore saw one of the most sustained high levels of investment in GDP in any country in world history with more than 30% of GDP devoted to fixed investment for 30 years from 1970-2000. Singapore’s Gross Domestic Fixed Capital Formation peaked at 47.4% of GDP in 1984, remained at 38.7% of GDP in 1997 and fell to 24.9% of GDP in 2007.

Singapore’s annual rate of GDP growth over the same period fell from 8.5% a year in 1984, and 9.7% a year in 1997, to 7.1% a year in 2007 (see Figures 9 and 10).

Figure 9

Figure 10

Hong Kong

The percentage of Hong Kong’s GDP devoted to Gross Domestic Fixed Capital Formation, amid significant fluctuations, fell from 35.6% in 1964 to 35.6% and to 20.3% in 2007.

Hong Kong’s annual average growth rate fell from 10.5% in 1964 to 6.4% in 2007 (see Figures 11 and 12).

Figure 11

Figure 12

China and India

India and China show a clear contrast to Japan, South Korea, Singapore and Hong Kong.

Whereas in Japan, South Korea, Singapore and Hong Kong there was a decline in the proportion of the economy devoted to investment and a decline in the rate of economic growth, both India and China India have systematically increased the share of investment in their GDP and have seen an acceleration of their growth rates. Because this pattern in India and China is so strikingly different to Japan, South Korea, Singapore and Hong Kong it is worth looking at in some detail.

India’s Gross Domestic fixed Capital Formation increased from 17.9% of GDP in 1977 to 22.7% of GDP in 2000 and to 33.9% of GDP in 2007. By the third quarter of 2008, before the onset of the international financial crisis, India’s Gross Domestic Capital Formation reached 37.6% of GDP. Over the same period India’s annual GDP growth rate accelerated from 4.5% in 1977 to 6.0% in 2000 and to 8.8% in 2007.

Unlike those who advocate a reduction in investment and savings rates, Manmohan Singh, who is not only India Prime Minister but an excellently trained economist, has constantly stressed the need to raise India’s savings and investment rates and has made this a foundation of his economic policy – with considerable success, as has been seen, in terms of sustaining high growth rates.

Manmohan Singh considered China’s high savings and investment rates as the foundation of superior economic performance. For example in 2003 when asked, ‘is it legitimate to compare India and Chinese economies?,’ he replied: ‘There is nothing wrong in the comparison. It is good to try and achieve the growth rate of China. But we must remember that the Chinese savings rate is 42 per cent of the Gross Domestic Product, whereas savings in India is hovering at 24 per cent.’

Before he became Prime Minister in May 2004 Singh set out clearly the investment rate without which India’s target growth rate could not be achieved: ‘at a Delhi seminar, Dr Manmohan Singh spoke out regarding the targeted eight per cent growth rate in the Tenth Plan... he opined that an eight per cent growth rate would require a 30 per cent ratio of savings to income and a substantial rise in the tax-GDP ratio.’

Therefore in 2006, after assuming office, Prime Minister Singh noted with satisfaction the increase in India’s savings rate and set the goal of increasing it further together with a concomitant rise in the the investment rate: ‘Our statisticians now tell me that our savings rate has shot up in the last couple of years to about 27 to 28 percent of our GDP… we are a country where the proportion of young people to total population is increasing. All demographers tell me that if we can find productive jobs for this young labour force, that itself should bring about a significant increase in India's savings rate in the next five to ten years. If our savings rate goes up, let us say, in the next ten years, by 5 percent of GDP, we would have generated the resources for investment in the management of this new urban infrastructure that we need in order to make a success of our attempt at modernization and growth.’

By 2007 Prime Minister Singh therefore welcomed the further increase in India’s savings and investment rates. According to India’s premier financial paper, the Economic Times: ‘The investment and saving rate is as high as 35 percent of national economic output, Singh said at a meeting of his Congress party in this southern Indian city, the hub of a 50-billion-dollar IT industry at the vanguard of the country's economic resurgence.’

Similarly, India’s finance minister, P Chidambaram , called in February 2007 for a further increase in India’s savings and investment rates: ‘India’s savings and investment rate as percentage of GDP have gone up by 2 per cent each. But to sustain the revised growth rate of 9 per cent in the 11th Plan, he [ P Chidambaram] said: “Both savings and investment as proportion of GDP must be raised further.”

By February 2008 Prime Minister Singh noted the continued advance of the savings rate and the new high reached in India’s investment rate: ‘Highlighting the strong fundamentals of the economy, Dr. Singh said that the savings rate in the country has touched almost 35 per cent of Gross Domestic Product (GDP) and the investment rate is at an all time peak of over 36 per cent of the GDP.’

The orientation of India to very high savings and investment rates, and the relation of this to rapid economic growth, is therefore clear (see Figures 13 and 14)

Considering China its fixed investment increased from 27.8% of GDP in 1978 to 34.3% of GDP in 2000 and to 42.7 % of GDP in 2007 42.7%. In the same period China’s rate of GDP growth accelerated from 4.9% in 1978 to 8.6% in 2000 and to 10.8% in 2007.

Figure 13

Figure 14

Conclusion

The conclusion from economic evidence is therefore clear.

A high percentage of fixed investment in GDP is an indispensible precondition for a rapid rate of growth – there are no examples of countries with rapid rates of GDP growth and low proportions of the GDP devoted to fixed investment. It is a high level of investment in GDP, not a high rate of consumption, that is necessary for rapid GDP growth.

In those countries in which the rate of investment in GDP fell – Japan, South Korea, Singpore and Hong Kong – the rate of economic growth also fell substantially. In those countries – India and China – in which the percentage of GDP devoted to investment rose the rate of economic growth also increased.

In short all evidence establishes clearly that it is the high rate of investment which is decisive for rapid GDP growth.

This overwhelming factual evidence, of course, supports what is evident from a theoretical point of view. Consumption, by definition, does not add to productivity potential or production capacity and therefore increasing the rate of consumption does not raise GDP growth. If China lowers its proportion of the economy devoted to investment its economic growth rate will also fall - as is confirmed by the international experience noted above.

* * *

The paper draws on earlier material which appeared in 'Why Asia will continue to grow more rapidly than the US and Europe - a historical perspective' on the blog Key Trends in Globalisation.

References

[1] The figure for England for 1688 is that in Angus Maddison, The World Economy, OECD Paris 2006 p395. UK figures after 1688 and up to 1947 are calculated from One Hundred Years of Economic Statistics, The Economist, London 1989 p74. Figures from 1948 are calculated from International Monetary Fund, International Financial Statistics (August 2008) Minor adjustments have been made to chain the earlier statistics to be consistent with the IMF data – in no case does this make any significant difference to the pattern shown. The data for fixed investment for the earlier period used by The Economist One Hundred Years of Economic Statistics are based on calculations in C H Feinstein and Pollard Studies in Capital Formation in the United Kingdon 1750-1820, Oxford University Press, Oxford 1988. Other commentators have suggested that Feinstein and Pollard's figures are somewhat too high - see for example. N F R Crafts British Economic Growth during the Industrial Revolution, Clarendon, Oxford 1986 p73. None of these revisions and differences however is of sufficient magnitude to alter the fundamental pattern shown here.
US figures prior to 1948 are calculated from One Hundred Years of Economic Statistics, The Economist, London 1989 p74. Figures from 1948 are calculated from International Monetary Fund, International Financial Statistics (August 2008) Data for the earlier period give only private fixed capital formation whereas that after 1948 is for total fixed capital formation – i.e. including government fixed capital formation. There are no reliable estimates of government fixed capital formation in the earlier period and therefore data for the earlier period have been adjusted upward by the difference between the two in 1948 – which is slightly over two per cent of GDP. This has the effect of revising upwards slightly the percentage of GDP allocated to fixed investment in the earlier period but the difference is too small to affect the overall pattern.
Figures for Germany prior to 1960 are calculated from One Hundred Years of Economic Statistics, The Economist, London 1989 p202. Figures from 1960 are calculated from International Monetary Fund, International Financial Statistics(August 2008). There is however no significant statistical difference between the two.
Figures for Japan, South Korea, China, India and Vietnam calculated from International Monetary Fund, International Financial Statistics.

[2] Phyllis Deane and W A Cole in British Economic Growth 1688-1959, Cambridge University Press, Cambridge 1980 p2 being closer to the lower figure while further studies have tended to revise the figure upwards slightly. The higher estimates for the earlier period have been taken here so as to avoid any suggestion of exaggerating the degree to which the proportion of GDP devoted to Gross Domestic Fixed Capital Formation has risen. The precise figure used here is that calculated by Maddison in Angus Maddison, The World Economy, OECD Paris 2006 p395. The higher figure, as can be seen, makes no difference to the overall trend.

[3] These countries are Iran, Iraq, Equatorial Guinea, Kuwait, Israel, Jordan, Oman, Qatar, Saudi Arabia, Libya, Gabon, Equatorial Guinea, and the United Arab Emirates.

Perspectives for China's exports

Summary

1. The Chinese authorities have rightly taken the strongest possible stance against protectionism.

2. While such a stance is correct it is necessary to have an accurate estimate of the objective evolution of world trade as far from all elements are under the control of China itself. The chief conclusions which emerge from such a study are serious and clear.

3. The current decline in international trade is of much greater dimensions than any seen since World War II. The only scale of comparison approaching it is 1929. As regards the current speed of decline of trade, although not, of course, as yet in its duration, the current fall in trade is so far actually more rapid than at the onset of the Great Depression after 1929.

4. Given the magnitude of this trade decline there is no justification for any view that there will be a short term recovery of trade or any recovery of the world economy in the second half of 2009.

5. Based on present trends a favourable estimate of the time it will take for world trade and China's exports to recover to previous levels is three years. However the situation is continuing to deteriorate and this timescale may be too short. This minimum three year recovery period must therefore be included in forward planning.

6. The severity of the decline in trade means that contingency planning for all countries, including China, must therefore be based on examination of more unfavourable, and not more favourable, scenarios than those currently envisaged.

The reasons for these conclusions will become clearer if the present situation of world trade is briefly placed in the more general framework of the unfolding of the international financial crisis.

The Chinese authorities correct stance against protectionism

The Chinese authorities have rightly taken the strongest possible stance against protectionism since the beginning of the international financial crisis. This consists not only of positions at international meetings, and in public statements, but practical measures such as the dispatch of trade delegations, simplifying procedures for Chinese companies to invest abroad, boosting trade with Taiwan, making possible trade in RMB with Hong Kong, and similar steps. Such policies will clearly continue to be necessary.

While such a stance is correct far from all elements of the international economic situation are under China's control. It is therefore necessary to have an accurate estimate of the objective evolution of world trade. Other countries may simply be overwhelmed by objective factors, given the severity of international financial trends, or lapse into protectionist policies. This article, therefore, considers current declines in China's trade in an international context and placing it in a comparative historical context. The chief conclusions which emerge are serious and
clear.

The current decline in international trade is of a severity that substantially exceeds those of 1973 - following the first oil price increase crisis,1981-82 - during the US Volker-Reagan recession, or in the 1997 South East Asian devaluation and debt crisis - i.e the current decline in trade is of much greater dimensions than any seen since World War II. The only scale of comparison approaching the current decline of world trade is 1929. As regards the current speed of decline of trade, although not of course as yet its duration, the current fall is actually more
rapid than in the onset of the Great Depression after 1929.

Current declines on international financial markets

In order to assess the current trends in the international economy it is important to understand the real dimensions of the decline in financial markets. To do so Figure 1 compares the decline in the Dow Jones Industrial Average during the current crisis with that following 1929 - i.e in comparison with the most serious financial crisis so far in economic history. The graph shows the percentage decline of the Dow Jones Industrial Average compared to the peak in the preceding cycle, in both 1929 and 2007, measured against the number of trading days since that peak. As may be seen the rapidity of the decline since 2007 almost exactly follows that after 1929.

Figure 1

In order to illustrate that this current rate of decline in share markets is in no way typical of a cyclical decline Figure 2 also shows the similar rates of decline of share prices during the other two major stock market falls of the 20th and 21st centuries - i.e following the oil price increase of 1973 and following the bursting of the dot com bubble in 2000. As may be seen the share price declines after 2007 and after 1929 far exceed the falls in 1973 and 2000, making clear that the only suitable point of comparison as regards financial markets for the current crisis is with 1929.

Figure 2

It should also be noted that, as shown in Figure 3, the rise in share prices on Wall Street in recent trading did not break out of this declining trend. The shift so far has simply moved the rate of descent of US share prices closer to the declining trendline that has been operating since October 2007 after several weeks of even more rapid than average falls.

Figure 3

The situation of world trade

Having established that current falls on share markets greatly exceed in scale any seen since World War II it may be rapidly established also that current trade declines also far exceed in scale any seen since World War II. Furthermore, as will be seen, trade plays a major role in assessing the scale of the present economic crisis in regard to the relation between the speed of decline in financial markets and the development of the situation in the productive economy.

Whereas present rates of fall on financial markets are comparable to 1929 most published data on the productive economy indicates a lesser rate of decline – i.e. there is a disparity between the two indicators. An exception is the annualised figures for US GDP for the 4th quarter of 2008 for investment, exports and imports which do indicate rates of declines of post-1929 dimensions. However, the annualised figure for US GDP itself for the 4th quarter of 2008
indicates a rate of economic contraction about two thirds as fast as in 1929.

However part of this apparent difference between financial and productive economy indicators can be that while financial markets may be followed in real time there is a delay in the calculation and publication of statistics on the productive economy. Current rates of economic decline are sufficiently rapid that delays between the actual current situation in the productive economy and the publication of data regarding it may lead to significant distortion in the picture presented in some statistics.

This is most serious in the case of GDP data. This is, first, based on data over a whole quarter and second it is necessarily published with delay due to the time taken to calculate it. Revised data for US 4th quarter GDP figures only became available at the end of February and, as monthly GDP data is not calculated it did not indicate the speed of change during the quarter itself - it is known from other measures that the rate of economic decline was accelerating significantly during the quarter. For that reason other indices than GDP, which are available more rapidly and on a more frequent basis, may indicate more accurately, if more partially, actual trends. Industrial production data, however, which is frequently taken as a more recent measure of economic shifts may be misleading in countries which are heavily dominated by
service sector activities.

One of the most important of available data sources is trade. This furthermore relates directly to one of the key issues in the current economic downturn – globalisation and protectionism. During
periods of severe recession a process of trade ‘de-globalisation’ frequently occurs – i.e. exports and imports fall more rapidly than GDP and therefore trade contracts as a percentage of GDP. The most extreme case of trade ‘de-globalisation’ was following 1929 - when the decline in exports and imports was far greater than that for GDP in the US and other economies and was exceeded in magnitude only by the fall in investment.

A similar pattern of the falling share of trade in GDP, on a much smaller scale, was seen during the most severe post-World War II recessions – for example in the US in 1981-82.

The scale of the current decline in world trade

To assess current trends in world trade Figure 4 shows the year on year decline of exports for Organisation for Economic Cooperation and Development (OECD) member countries. As these account for the majority of world trade, and the main non-OECD economies such as China are considered below and show the same pattern, these trends clearly reflect the overall situation of world trade.

The trend shown in Figure 4 is clear. The year on year decline in OECD exports to December 2008 was 18.9 per cent - the most serious in the period since World War II.

Figure 4

Moreover world exports were declining sharply during the last three months of 2008, after having risen strongly, at the beginning of the year. The year on year comparison therefore understates the rapidity of the decline in trade at the end of the year. To illustrate this latter trend Figure 5 therefore shows the fall in exports for the last three months of 2008 on an annualised basis. The 64.5% annualised rate of decline of OECD exports for the last three months of 2008 far exceeds any fall seen in post-World War II history.

Figure 5

Given that the current decline in world trade considerably exceeds in magnitude any seen since World War II comparisons only with previous post-World War II recessions are misleading in that they significantly understate the dimensions of the fall.

Taking as a benchmark the most serious trade decline in world history, that following 1929, casts the magnitude of the present situation in a clearer light. Such comparison indicates that, as regards a whole series of countries, the current decline in trade is more rapid than that following 1929. If that process were to continue for any significant period of time the consequences, both in terms of the recessionary pressures that are implied and in terms of ‘de-globalisation’ and the rise of protectionist trends, would be extremely significant. These trends in trade, therefore, also provide some evidence that the deep pessimism in financial markets, with post-1929 speeds of decline, may be justified by events in the productive economy.

To give an initial benchmark Table 1 shows the decline in exports for countries for which data is available both for the total decline in exports in the period 1929-33 and in the first year
following 1929. The data is measured in current prices to make it comparable with the most recent data for the decline in exports from 2008 onwards. As may be seen all countries suffered major export declines in 1929-30 - the most severe contractions being in the UK, Canada, the US and Japan with, respectively, annual falls in exports in 1929-30 of 19.3%, 20.9%, 22.5%, and 28.1%. The US also suffered the most severe decline for the period 1929-33 as a whole - with a total decline in exports over a four year period of 66.2%.

Table 1

Current declines in world trade

The latest figures released by the Organisation for Economic Co-operation and Development (OECD) for world trade up to December 2008, with data for more recent months in the case of certain states, allows the calculation of a relatively comprehensive picture regarding world trade for the major economies. The pattern is extremely striking. The rates of decline of world trade in many case are more rapid than in 1929 and in a number of cases the decline in trade since 2008 is already approaching the declines for the 1929-33 period as a whole. In short the contraction of world trade is today of extreme rapidity.

Due to the extremely rapid shift in the situation three indicators have been calculated for exports – the actual year on year decline to December 2008, the actual decline in exports since the peak month for each country or area last year, and the change during the three months to December 2008 on an annualised basis.

In order to give a year by year historical scale of comparison the decline of US exports, in current prices, was 22.5% in 1929-30, 32.7% in 1930-31, 32.4% in 1931-32 and 4.0% in 1932-33 after which partial export recovery commenced - i.e. the most rapid annual rate of decline of US exports in the Great Depression, and the most rapid on record to date, was 32.7% in 1930-31. By 1933 US exports, as noted in Table 1, had fallen 66.2% below their 1929 level.

Considering first the OECD area as a whole, Table 2 sets out in more detail data already illustrated in Figures 4 and 5 and also analyses the situation in the OECD European region. As can be seen, for the OECD region as a whole exports have already declined by over 30% since their peak in April 2008 - essentially equalling the rates of decline of the worst year of the 1930s and exceeding the decline of US trade in 1929-30. The annualised rate of decline in the three months up to December 2008 was 64%.

For the major G7 economies the decline was only slightly less severe - with a decline of 26.9% since the peak in July 2008 and an annualised rate of decline of 57.8% in the three months to December 2008.

Within the Euro area the annualised rate of decline for the three months to December 2008 was 50.4% and for the OECD European region, which includes some East European states, the annualised rate of decline was 67.0%.

It may therefore be clearly said that in the field of trade, as in that of financial markets, the current decline is full comparable in speed of descent to the onset of the Great Depression. The difference, so far, is not in the speed of fall but in its duration - the decline in exports after 1929 continued for four years whereas so far the current decline has been occurring for a year. It remains to be seen for how long the current decline with continue.

Table 2

Turning to individual countries, Table 3 shows the figures for the largest OECD economies - the G7. As may be seen all G7 countries have seen declines in exports of over 25% since their peak levels last year and in the three months to December 2008 all witnessed annualised rates of decline of more than 50%.

In short, the precipitate decline in world trade, at 1930s rates of descent, is not confined to smaller economies but fully affects the largest ones. The actual decline in trade in all G7 countries since the peak in 2008 is greater than the fall in the US in 1929-30.

Table 3

Table 4 shows the rates of decline of exports for the non-G7 European OECD states. As may be seen with the exception of two small economies, Luxemburg and Ireland, which have done better than others, all OECD European countries have seen actual export declines of at least 25% and annualised rates of decline of 50% or more. Again the actual decline in trade since the peak in 2008 is already greater than for the US in 1929-30.

It is possible that the rate of decline for Spain, an incredible 99.7% annualised rate in the three months to December 2008, is a statistical freak or error but the annualised rates of decline for Sweden, Poland, and Norway are almost as severe - respectively 79.1%, 82,8%, and 83.1%. Such rates may rightly be characterised not as decline but of collapse of exports in at least the short term.

Table 4

Turning to non-European economies, the data is set out in Table 5. Again, with the exception of the small economies of Iceland and New Zealand, the highly publicised decline of Chinese exports by 22.3% since their peak last year year, and at an annualised rate of 53.0% in the three months to December, are themselves actually significantly smaller than for other countries. Mexico and South Korea have already seen actual declines of exports of over 30% and South Africa and Turkey have seen falls of over 40%. The annualised rates of decline of exports for South Korea, Brazil, Indonesia, South Africa, and Turkey - at 70.7%, 72.4%, 78.2%, 82.1%, and 90.1% respectively - are clearly catastrophic.

Table 5

Countries for which OECD data is available for January confirm continuation of the same trend – as shown in Table 6. The chief difference is that with the extra month the actual declines in exports, as opposed to only the annualised rates of fall, have become more serious.

The actual falls recorded from the maximum levels of exports are 29.8% for Switzerland, 41.1% for South Africa, 41.4% for Sweden, 46.3% for Norway and 47.5% for Turkey. There is nothing in this pattern which indicates results for other countries are likely to show an improved tendency.

Table 6

Summary of the current rends in world trade

Summarising the above data, of the 34 countries studied 14 had annualised rates of decline of exports of more than 70% and 20 had rates of decline of more than 60%. The widely publicised reports of declines of exports in the last three months of last year such as the annualised 51.9% for Japan, 53.0% for China, or 54.0% for the US, which attracted much publicity, are actually smaller than the falls in most countries.

While the annualised rates of decline show the extremely striking implosion of world trade during the last three months of 2008 an annualised rate, naturally, indicates an, in this case extremely severe, tendency. What is equally disturbing is the actual falls in exports recorded from the maximum levels last year. Seven countries registered actual falls in exports of more than 40% and 19 of more than 30%.

The relation of the decline in financial markets, trade and the productive economy

The transmission mechanisms of the financial crisis into the productive economy are also made clear by the trends outlined above. Initially in the present crisis there was a disjunction between the decline in financial markets, which was of 1929 magnitude, and the situation of the productive economy - which was of a severe but not equivalent decline. As such a disjunction is highly unlikely to continue either financial markets would recover, having overshot on the downside, or the trends and statistics in the productive economy would be shown to have been a lagging indicator and they would adjust downwards to the tendencies indicated in financial markets.

The extraordinarily powerful falls in world exports shown in the latest figures for all major economies indicate that the decline in trade is operating as a key mechanism by which the crisis revealed in financial markets is beginning to affect the productive economy. It may now be said that in two areas of the world economy, financial markets and trade, rates of decline significantly exceed any seen since World War II and are fully comparable to 1929 scale. How powerful the transmission mechanisms from the international sector are into domestic economies must clearly be carefully studied - trade today plays a more significant role in the world economy than at the onset of the 1929 crisis.

The duration of the crisis is also critical - the so far unique severity of 1929 was not only due to the rapidity of the fall but by its duration. The decline in US trade and GDP in the 1930s continued for four years whereas the current decline in financial markets has lasted 17 months, the decline in trade slightly under one year, and the fall in GDP approximately six months.

Quite sufficient data are now in to say with certainty that in the last three months of 2008 a convulsion in world trade occurred. The extreme rapidity of the fall in world trade, as with the situation in financial markets, confirms that the benchmark for present analyses must be not only, or primarily, post-World War II recessions but also 1929 itself.

Perspectives for China's exports

The above international situation provides the framework for estimates of perspectives for China's exports.

As may be seen from the figures above China's export decline is not exceptional by the standards of current trends in trade - on the contrary China's decline in exports is moderate compared to most countries. This makes it extremely unlikely that the decline in China's decline in exports seen since last year is due to some exceptional factor that can be easily reversed. On the contrary China's decline in exports is inextricably tied to trends in the world economy and depends on recovery in the world economy.

Figure 6 shows the trend in OECD country exports in dollar terms - Figure 7 shows the same data graphed on a log axis to show the rate of change. Both confirm that the decline in international trade since April 2008 is without precedent since World War II - the period covered includes both of the severe post-war trade crises in 1973 and 1981. As this unprecedented decline is not rooted in specific policies pursued by China it cannot be reversed by specific policies adopted by China.

The logarithmic graph also shows that there are no major 'leaps' in the growth of world trade i.e periods of extraordinarily rapid expansion. This is crucial as it indicates that recovery from any decline in trade cannot be immediate - it is rooted in the periods of time necessary to develop trade volme. Therefore, given the magnitude of the decline in exports, recovery from the present extremely severe decline in world trade will necessarily not be rapid.

Figure 6

Figure 7

This development of OECD trade also gives a framework for assessing China's trade - which is itself shown in Figures 8 and 9. To show both the absolute shifts and the rates of change Figures 8 and 9 again use both linear (Figure 8) and logarithmic (figure 9) axes. The again show that there are no 'leaps' in the expansion of trade - i.e. any recovery will take a significant period of time (technically R² is greater than 0.9 for the correlation between time and trade value).

Figure 8

Figure 9

In order to assess how long it will take China's exports to recover it may be noted that by January-February 2009 China's exports had declined to the level of the same period in 2006. Given that there is no evidence in the long term data of any possibility of radical upward 'leaps' in trade for either the OECD or China a reasonable working hypothesis would be that it will take China's exports at least three years to regain the level of August 2008.

However, it should be noted that the situation of world trade is continuing to deteriorate. The three year recovery period for China's exports flowing from this international data to December 2008 may well be, therefore, too optimistic. This again confirms that the working hypothesis should be that, based on trends within world trade, it will take at least three years for recovery of China's exports to their previous levels.

The implications for China's companies, economy and domestic stimulus programmes of the above trends are evident - if it is estimated that at least three years will be recovered for China's exports to recover to their levels of summer 2008 then planning for all three must take place on that basis.

Conclusion

Analyses which focus on comparisons with trade declines since World War II are too optimistic as the present decline in world trade far exceeds in its dimensions any seen since 1945. The present rate of decline of world trade is actually more rapid than in 1929 - although, of course, the duration of the fall is so far, of course, far shorter. Given such trends any perspective of rapid recovery of China's exports is excluded. The working scenario should be that it will take at least three years for China's exports to recover to their previous levels. If current trends of decline of world trade continue that period would have to be lengthened. Company and economic planning should, therefore, take place on that basis.

Notes to Tables - peak month for exports in 2008

1. Peak January 2008
2. Peak March 2008
3. Peak April 2008
4. Peak May 2008
5. Peak June 2008
6. Peak July 2008
7. Peak August 2008
8. Peak September 2008

美国四季度GDP数据显示1929时期的衰退速度

媒体关于官方公布的第四季度的美国GDP数据的报告主要集中于GDP年下降比率的修改上，由最初的6.2%调整为现在的3.8%。报告强调这个下降比率比独立预测的平均下降5.4%严重的多。

然而，集中分析新闻标题中GDP数据，使得人们不去分析GDP组成成分的具体趋势。而其中很多部分又特别重要。

就像在全球化过程中的核心趋势中分析的那样，现在的国际金融危机中的明显的‘矛盾’就是金融市场的衰退速度和1929年的大致相仿，而实体经济中的衰退却没有1929年那么严重。

这样的矛盾是不大可能一直保持的。一种情况就是，金融市场过度反应了，衰退的过度了。生产将不会像1929年那样下降，金融市场也将恢复过来。这样的话，只会出现经济衰退，而不会出现大萧条。另一种情况就是，现在的金融市场的1929年程度的衰退是合理的，实体经济的1929年程度的衰退只是被推迟了，并将在不久后大幅下降。这样的话，经济就不是简单的衰退，而是走向大萧条。

美国4季度GDP数据的重要性在于，GDP各成分的下降预示着第二种情况更有可能发生。
这样的趋势，并不一定预示着美国经济就会出现和大萧条时期一样的GDP衰退-GDP下降29.7%。但是这预示着比二战后任何时期都大得多的衰退。在二战后最严重的衰退是1981-1982年的1.9%。

根据最新的数据，在第四季度，美国GDP以每年6.2%的速度下降。消费支出年下降4.3%，私人固定投资年下降20.8%，出口年下降23.6%，进口年下降16.0%。

美国投资的下降不局限于由于次贷危机影响住宅部门投资的下降。住宅部分投资下降22.2%，只比全部投资的下降比率高出一点点。

为了更好的理解这些数据，最好是将它们与1929-30年大萧条刚开始时的数据进行比较。在那年，GDP下降9.4%，消费支出下降6.7%，私人部分投资下降23.4%，出口下降15.4%，进口下降11.0%。

比较这两组数据可以看出，2008年美国GDP的下降大概是1929-1930年的三分之二，消费支出的下降也大概是大萧条开始第一年的三分之二。但是2008年第四季度投资的下降和1929-30年几乎一样严重，出口和进口的下降则比1929-30年更糟。

做这些比较需要注意的是，将第四季度的数据年度化以后的下降，并不能完全用来和1929年发生的真实的下降进行比较。还需要观察2009年美国经济的真实下降。然而2008年第四季度美国投资、进口和出口的下降已经完全可以和1929-30年相当了。

其它像1929年那种程度的下降的国际数据包括亚洲交易和工业生产水平的快速下降。比如2009年一月日本出口同比下降45.7%，工业生产同比下降30.0%。这些数据无疑会由于中国春节期间亚洲地区经济下滑进一步恶化。这样的下降完全和1929年相当。

总之，金融市场下滑速度和实体经济的下滑速度不匹配的矛盾正被一些国家实体经济的快速下滑所解决。这是不好的情形，像1929年，一开始实体经济就快速下降。这种趋势只是出现了，但是还没有被证实。然而这种趋势的出现意味着政府、政策制定者和企业不能只试图减少正在发生的事情的严重性。为最坏的情形做好准备是必不可少的。

只谈衰退是会产生误导的。应该讨论的是将要发生的是二战后最严重的衰退，还是一场即将发生的大萧条。

* **
这篇文章最初发表在博客： Key Trends in Globalisation.

4th Quarter US GDP figures show shift to post-1929 rates of decline

Media reports on the latest release of official 4th quarter US GDP figures have concentrated on the headline revision of the annualised rate of decline of GDP from 3.8%, in the first published estimate, to 6.2% in the new one. It has also been stressed that this decline was significantly worse than the average of independent projections - which was for a 5.4% annualised fall.

However concentration on the headline GDP figure has distracted from examination of the detailed trends of the components of GDP, some of which are of very significant magnitude.

Until now the apparent ‘paradox’ of the international financial crisis was that, as Key Trends in Globalisation has analysed, the decline in financial markets was entirely comparable in rapidity to 1929, however the decline in the productive economy was far less severe than after 1929.

Such a paradox is highly unlikely to be maintained. The first possibility is that it will be shown that financial markets had overshot, that their decline is excessive, that there will not be anything approaching a post-1929 scale fall in production, and consequently financial markets will adjust back upwards – i.e. the variant that there will be an economic recession not a depression. An alternative is that it will be found that the 1929 scale financial falls are justified, and that it is merely a matter of a delay in time before the productive economy adjusts very severely downwards by the amounts that would justify a 1929 scale financial fall – i.e. the perspective not simply of a recession but of something approaching an economic depression.

The significance of the details of the 4th Quarter US GDP figures is that the rates of decline of components of GDP indicate that there is now a higher possibility of the second variant.

Such a trend does not, of course, necessarily imply that the scale of decline in US GDP will be the same as in the Great Depression - when GDP fell by 29.7%. But it would mean falls far exceeding in magnitude any post-World War II recession – the most severe recession in post-war US history being 1981-82 when GDP declined by 1.9%.

According to the new data, in the 4th quarter of 2008 US GDP declined at an annualised rate of 6.2%, consumer expenditure declined at an annualised 4.3%, private fixed investment fell at an annualised 20.8%, exports declined by an annualised 23.6%, and imports dropped at an annualised 16.0%.

The decline in US investment was not accounted for solely by the fall in the residential sector, produced by the sub-prime mortgage crisis, as the latter dropped by 22.2% - only marginally more than the overall investment decline.

To grasp the scale of magnitude of such rates of decline it is worth making a comparison to the actual falls in 1929-30 in the US – i.e. in the first year of the onset of the Great Depression. In that year US GDP fell by 9.4%, consumer expenditure by 6.7%, private fixed investment by 23.4%, exports by 15.4% and imports by 11.0%.

Comparing the two sets of figures, the annualised decline in US GDP in the fourth quarter of 2008 was about two thirds as rapid as the actual yearly decline in 1929-30 and the fall in consumer expenditure was similarly about two thirds of the first year of the Great Depression. But the rate of fall in investment in the 4th Quarter of 2008 was almost as severe as in 1929-30 and the rate of fall in US exports and imports was actually worse than in 1929-30.

Having made such a comparison it is necessary to state immediately that an annualised rate of decline is not the same as an actual annual rate of decline which has occurred as in 1929. It remains to be seen what will be the actual drop in US GDP during 2009. However the rates of decline of US investment, exports and imports in the last quarter of 2008 were fully comparable to those in 1929-30.

Other international data which goes in the same direction of post-1929 scales of fall are the extremely rapid rates of decline in Asian trade and industrial production. Exports by Japan, for example, dropped by 45.7% in January 2009 compared to a year earlier and its industrial production was down 30.0% over the same period. While these figures are undoubtedly exacerbated by the slowdown in the whole Asian economic region caused by the Chinese lunar New Year festival falling unusually early this year, in January, nevertheless anything approaching such scales of decline are figures fully comparable to 1929.

In short, elements are now appearing in the productive economy of a number of countries which overcome the’ paradox’ of the gap between the rate of decline of financial markets and the rate of decline of the productive economy in the negative variant – i.e. by the onset of declines in the productive economy of post-1929 magnitudes. Such trends are, of course, only just appearing and as yet are not consolidated. However the appearance of such tendencies clearly means that governments, policy makers and companies should not be seeking to minimise the gravity of what is taking place. It is more imperative at present to prepare for worse case scenarios than optimistic ones.

Talk simply of ‘recession’ is misleading. The issue is whether what will take place will be ‘merely’ the most serious recession since World War II or whether an actual economic depression will set in.

* * *

This article originally appeared on the blog Key Trends in Globalisation.

US share price decline continues to match fall after 1929

The latest market shifts confirm that to date in the present financial crisis the fall in US share prices continues to be at a post-1929 rate – see Figure 1.

Figure 1

Last Friday, 27 February, was the 352nd trading day since the all time high of the Dow Jones Industrial Average on 9 October 2007. Since its peak the Dow has declined by 50.1%. On the 352nd trading day following its 1929 high point the Dow had declined by 55.6%. The difference between the falls is within the range of short term fluctuations. Overall the rate of decline of US share prices since October 2007 is tracking the 1929 descent.

As may be seen from Figure 2 the current decline of share prices is far more rapid and deep than either of the other two major falls in the 20th century – that following the oil price increase of 1973 or following the implosion of the dot com share bubble in 2000. These trends confirm again that the only relevant scale of comparison for the current decline of US share prices is with 1929 itself.

Figure 2

The only fundamental difference between the current decline and that of 1929, so far, is the duration of the fall. After 1929 it took the Dow Jones 713 trading days to reach its bottom – the trough being on 8 July 1932 by which time the Dow it had lost 89.2% of its value. It remains to be seen for how long the current decline will continue. However although the duration of the drop is at least as yet not as great as after 1929 it is as rapid.

While the financial collapse is therefore genuinely on a scale to be compared to 1929, the overall declines registered in the productive economy are still far smaller than the post-1929 falls. However the declines in the productive economy have just commenced and data is still coming in. Some drops, such as in trade, are approaching scales seen post-1929. However the decline in US GDP in the 4th quarter, now revised downwards to an annualised decline of 6.2 per cent, is about two thirds of the 9.4 per cent drop in US GDP seen between 1929 and 1930 – however the rate of decline is accelerating.
While quite sufficient information is now in to make clear that the current financial crisis is only comparable to 1929, far exceeding the shifts seen in a normal recession, further data is still required to see whether it will be meaningful to compared the fall in production to that after 1929 or not.

Present rate of decline in world trade more rapid than in 1929 – by John Ross and Lu Yao

A crucial question is assessing the current scale of the international economic crisis is examining the relation between the speed of decline in financial markets and the situation in the productive economy.

As this blog has analysed on several occasions the rates of present falls on financial markets are entirely comparable to those following 1929 – i.e. the most rapid declines in history. Most published data on the productive economy, however, indicates a rate of decline not approaching that of 1929 – i.e. there is a disparity between the two indicators. An exception is that annualised figures for US GDP for the 4th quarter of 2008 regarding investment, exports and imports indicate rates of declines of post-1929 dimensions. However the annualised figure for US GDP itself for the 4th quarter of 2008 indicates a rate of fall about two thirds as fast as in 1929.

However, a part of this apparent ‘paradox’ of the disparity between financial and productive economy data can be that while financial markets may be followed in real time there is a delay in the calculation and publication of statistics on the productive economy. Current rates of economic decline are sufficiently rapid that delays between the actual current situation in the productive economy and the publication of data regarding it may lead to a significant distortion in the picture presented by at least some statistics.

This is most serious in the case of GDP data which is necessarily published with considerable delay due to the time taken to calculate it – revised data for US 4th quarter GDP figures only became available at the end of February. For that reason other indices than GDP, which may be available more rapidly, may indicate more accurately, if more partially, actual trends. Industrial production data, however, which is frequently taken as a more immediate measure of economic shifts, may be misleading in countries which are heavily dominated by service sector activities.

One of the most important of the available data sources is trade. This also directly relates to one of the key issues in the current economic downturn – globalisation and protectionism. During periods of severe recession a process of trade ‘de-globalisation’ frequently occurs – i.e. exports and imports fall more rapidly than GDP and therefore trade contracts as a percentage of GDP. The most extreme case of such trade ‘de-globalisation’ was following 1929 - when the decline in exports and imports was far greater than that for GDP in the US and other economies and was exceeded in magnitude only by the fall in investment. A similar pattern of the falling share of trade in GDP, on a much smaller scale, was seen during the most severe post-World War II recessions – for example in the US in 1981-82.

This article, therefore, calculates data on recent declines in exports and, as a benchmark, compares these to declines after 1929. The results are extremely striking. They indicate that, at least as regards a series of major countries, the current decline in trade is more rapid than that following 1929. If that process were to continue for any significant period of time the consequences, both in terms of the recessionary pressures that are implied and in terms of ‘de-globalisation’ and the rise of protectionist trends, would be extremely significant. These trends in trade, therefore, also provide some evidence that the extreme pessimism in financial markets, with post-1929 speeds of decline, may be justified by events in the productive economy.

To give an initial benchmark Table 1 shows the decline in exports for countries for which data is available in the first year following 1929. This data is the actual year on year decline measured in current price terms. As may be seen all countries suffered major export declines - the most severe contractions being in the US, Japan, Canada and the UK.

For comparison Table 2 shows the export trends to December 2008 for a series of major trading countries. December has been taken as the cut off point as Asia’s, and to a lesser extent other regions, trade figures are highly distorted by the fact that the Chinese lunar New Year holiday fell unusually early this year in January. This distorts year on year comparisons for January and February 2009.¹

It should be stressed, therefore, in the light of the magnitude of the shifts revealed, that as the economic situation has been deteriorating the figures calculated to December indicate the scale of falls conservatively. It is likely trends worsened in January and February.

To indicate the rate of deterioration three figures are shown in Table 2 – the actual year on year fall in exports to December 2008, the six month June-December 2008 decline on an annualised basis, and the three month September to December fall on an annualised basis.

It may be seen that the rates of decline of exports in the last half of 2008 were of extraordinary magnitudes – even considerably exceeding the rates of decline seen in 1929-30. For the US to have suffered annualised rates of decline of exports of 32.6% for the last six months, and for 44.5% for the last three months is, respectively, half as much again as, and almost twice, the decline seen in 1929.

The annualised rates of decline for the last three months in South Korea and Japan, 71.8% and 81.5% respectively, may well have no equivalents for peacetime in major countries in history. Even the annualised figures for the rates of decline for exports for the last six months for Germany, South Korea, and Japan – 42.0%, 46.3% and 54.5% respectively – are at speeds which considerably exceed those for 1929.

An immediate caution must be made that an annualised rate of decline, even one based on a six month period, is not the same as an actual annual decline. In most countries the decline in exports commenced in July 2008 (in March 2008 in Japan and in October 2008 in Germany) and therefore actual year on year results for the period since then will not be available for some months. But the scales of drop in the last six months are so sharp that there would have to be an extremely great, and in the economic circumstances highly unrealistic, recovery of trade in the next few months for the declines in exports on a year by year basis not to be at least as bad, and probably worse, than those for 1929-30 – in some cases worse than 1929-30 by a significant margin.

As a further control on the data to avoid exaggeration Table 3 shows the actual (i.e. non-annualised) decline in exports since the peak month for the countries concerned. The trends, again, are evident. In all cases the actual drop in exports that has already taken place since the maximum levels last year is essentially equal to or exceeds that seen in 1929.

Such rates of decline cast further light on the apparent differences between the post-1929 scale of falls on financial market and the apparently much less severe declines in the published statistics for the productive economy. Such extreme falls in exports/trade increase the probability that financial markets are behaving rationally and that at least part of the apparent disparity between the post-1929 scales of decline in financial markets and the lesser falls in published statistical data is due to the time lag in statistical calculation – i.e. that the actual situation in the productive economy is worse than it appears from the necessarily time lagged published data.

Second, it indicates that the 4th quarter US GDP data, which indicated an extremely severe and rapid deterioration in trade, is not exceptional but may be repeated for other countries.

Such figures indicate that in at least one area, trade, the rate of decline in the productive economy is now at least as rapid as after 1929 – i.e. the most severe ever recorded in peacetime history.

Notes

¹ For example South Korea’s export figures show a 32.8% year on year decline for January 2009 compared to a similar 17.3% year on year drop for February with this being likely to reflect not a rebound but depressed figures for January due to the holiday.

* * *

Antai College of Economic and Management at Jiao Tong University in Shanghai provided research facilities for this study.

谁是经济危机的幕后推手

世界经济正在遭受1929年以来最严重的金融危机，这一点已经无须赘言了。人们也开始对这次经济危机所产生的后果有了更深刻的认识。就美国而言，政府对于经济管制越放越宽的时代已经终结，取而代之的是大规模的政府干预经济。

但是这次的危机到底属于哪种类型呢？这很大程度上决定了这次经济危机的最终结果究竟会如何。

在新闻小报和煽动性的政治演讲中，提出最平淡无奇而肤浅的解释是，这是一场非常严重的短时紊乱，始作俑者正是那些“金融投机者”和“买空者”。这就象是上世纪60年代英国首相Harold Wilson指责“苏黎世侏儒”的现代版本。但这种解释至少要比某些金融机构普遍持有的主观而错误的判断要稍强那么一点。一个与此相关的观点认为心理因素才是导致危机的根源。因此，有一种普遍观点宣称这次金融危机的核心问题是“信心”或是“信心缺失”。

如果上述观点都是正确的话，那么即使目前的金融动荡很严重，但是其随后的调整应该是相对较小的。一旦目前的心理状态发生转变，也就是大家能够重拾信心，那么就能恢复到接近先前的水平，当然肯定也会有个别的变化。

然而，这样的分析是不正确的。当前的金融危机不是源于心理或是主观失误，而是源于其长期的经济发展趋势。所以其根源是客观的而不是主观的。因为心理因素并不足以推动客观经济，心理因素是随这客观经济的变化所产生的。

金融危机的不断演变体现出了要考虑长期趋势和潜在发展的重要性，而不要像很多新闻评论里那样仅仅是注重短期的变化或是断章取义的个别因素。 不能从全局看待问题就不能正确的理解前期的形势，这不可避免的导致对目前所处经济情况的困惑。

有关目前金融危机的真正驱动的力量的结果和结论将在这篇文章的结尾阐述。但首先，我们来罗列一些决定性的数据。

我在之前的文章《为何中国能继续比美国和欧洲更快速的增长》中曾经提到导致目前金融危机的最根本原因。这表明从投资经济占的比例来看，如今美国经济发展远远落后于主要的亚洲国家，其结果是美国经济的国际竞争力在不断下降。

大家可以从图表一可以看到这个趋势，这张图表曾在本博客之前文章里出现过。图表一显示美国、中国、印度的投资水平（国内固定资产投资总额）占其各自GDP的比例。美国的投资仅占其GDP的18-20%，而印度占30%，中国的投资甚至超过其GDP的40%。可以看到中国的经济以每年9%-11%的速度增长，印度的年增长是7%-9%，而美国仅为3%。

图表一

美国在投资方面呈现明显落后的态势。大家知道投资是提升其竞争力的关键因素。对于任一给定的汇率，美国经济正随着时间的推移逐渐的丧失其竞争力。如果美国不把它的投资水平提高到像其他经济体系那样，那么美国的竞争力还会继续下降，而保持其竞争力的唯一途径是不断降低汇率，也就是让美元逐步贬值。

图表二可以看到美国竞争力正在日益下降。众所周知，美国经受着巨大的国际收支逆差。这张图表显示的是其各阶段演变的过程。

图表二

美国的国际收支在80年代初期开始陷入大幅赤字，到80年代中期其赤字超过GDP的3%。美国的国际收支在90年代初有一个短暂的复苏，其原因是美元贬值，90年代初国内经济衰退，同时美国获得来自一笔来自海湾国家的钱，这些海湾国家的目的为了资助海湾战争。但这次复苏只是短暂的，1991年后美国的收支逆差变得更加严重，到2006年其收支逆差达到最高峰，超过GPD的6％，大约在每年7000亿美元左右。

图表三呈现了这种长期以来的发展态势。图表曲线表示了1956年来世界其他国家与美国之间的净借支。当这个数字为负值时，说明美国是在借入。这些趋势验证了美国经济竞争力的衰退。

图表三

美国的低投资意味着美国要保持其竞争力的唯一的基本机制就是美元贬值。图表四显示相对世界上其它三种非美元中的两种的货币-欧元、日元，美元汇率的一个长期走势。美元跟同一时期的第三种货币－人民币相比是没有意义的，因为直到20世纪80年代初中国一直都人为设定汇率，其用意就是不想参与世界贸易。

图表四

美元汇率经历了三个清晰的发展阶段:
- 从战后时期到70年初期，美国保持固定汇率的布雷顿森林货币体系。
- 从1973年直到80年代中期，美元汇率出现短期的大幅波动，实质性的美元贬值发生了。
- 从20世纪80年代中期以后，虽然有一些短期的波动，正式的浮动汇率机制已经开始实行了，事实上美元汇率呈现长期相对稳定的趋势。

已经如图表二所示，美国日益恶化的国际收支状况是80年代中期之后以下两种情况结合下的必然结果。

－美国较低投资率导致竞争力日益恶化

－稳定的美元汇率，即使采取贬值的措施也无法阻止竞争水平的下滑。

如图表二所示必定意味着美国国际收支逆差日益严重的趋势。换句话说，相对于美国潜在的竞争力，美元被逐渐高估了。

图表五显示，美元汇率保持稳定，尽管国际收支的日益恶化以及没有正式固定汇率。净国外购买的美国债务证券、国库债券和其他债券占GDP的比例从80年早初期的大约0%上升到2007年的5.6% ，大约是7700亿美元。国外国家对美国债务证券的大量投资使美国日益恶化的收支逆差得到平衡，也有利于维持稳定的美元汇率。

图表五

然而，想要欺骗任何经济力量是不可能成功的。所有有关经济的历史都告诉我们想要人为维持高汇率而不顾经济力量的压力最终还是会失败。所以，美元贬值会不可避免的发生。

高估美元的结果自然也就很清楚了。在国际环境中，由于美元的高估，这些用美元计价的资产价值也就必然被高估。因此，这些资产最终会价值回归到它们真实的水平，也就是更低的价值。一旦发生贬值，那么就会破坏或吞噬机构的资产负债表，从而引发美国的金融机构相继破产倒闭，这就是目前正在发生的情况。由于美元贬值和价值回归，其他两个的情况也就随之发生。

那些拥有美元资产的国外持有者损失惨重。单去年的流入的美国债务证券总计高达7700亿美元，其损失是巨大的。据估计日本大约持有8600亿美国债务证券包括国债和债券和拥有750亿美国最大的房地产抵押公司房地美和房利美的债券及担保证券，这两大房地产抵押机构已被美国国有化。中国持有美国大量的国债和高达4000亿房地美和房利的债券及担保证券。

美国资产的贬值也将引发美国各利益集团的政治斗争，谁将会承受这样的损失。

鉴于上述对美元的高估不可避免的导致金融危机的发生。目前美国的金融危机本质上就是其为了维持高估汇率所付出的代价，类似的危机也曾经在其他国家发生过，比如1982年墨西哥、1998年俄罗斯、2001年阿根廷都曾遭受此类危机。他们都曾因为企图维持高估汇率而最终都不可避免的溃败了。其国家金融机构也因此遭受了重创。而美国经济拥有更大的规模同时美国在全球经济发挥着至关重要的作用，这也导致了这次经济危机后果比以往都严重的多。

然而，如果这次的危机是相当典型的形式的话，那么就如本文开头概述的这次危机应该归于哪种类型呢？

前一时期很多评论家不考虑基本面和总体趋势，认为美国经济正在变的更有竞争力，以致于它创造了一个所谓的新经济。这种仅仅通过观察个别领域特别是高科技行业所得出的观点显然是不正确的。不可否认，美国在高科技领域的确具有很强的竞争力，但是，只要看一下整个美国国际收支情况，就可以看到美国经济正在不断的萎缩，而不是增强其竞争力。

所以金融危机并不是心理因素产生的，也不是在短期产生的，它是根植于一个长期客观经济的运行过程。事实上，可以说是政治因素决定了经济危机的演变和结果。

在以后的文章里，我还会阐述经济和政治因素之间的关联。但是导致目前金融危机的根本原因应该是非常清楚的，也符合上述的趋势。如果美国想要达到其亚洲竞争对手的投资水平，从短期来看，需要从消费到投资进行大规模的资源转移，这势必将会引起美国民众的不满从而导致政治局面不稳定。所以短期来看这种资源的转移应该不会发生。那么另外一个办法就是使其在短期内恢复竞争力，也就是让美元贬值。但是美元贬值会导致美元资产缩水，金融机构的资产负债表上表现为资不抵债，从而加剧金融的不稳定性。任何一种手段都会对美国民众的生活水平产生巨大的压力，因此也会对美国政治有巨大的影响。

* * *

本文《谁是经济危机的幕后推手》最初发表在Key Trends in Globalisation.

参考[1] New York Times指出美国正遭受自1929年经济大萧条以来最严重的经济危机。

参考[2] Wall Street Journal 指出美国资本主义在金融动荡中正在建立新的规则。更准确的说：“这场自1929年经济大萧条之后出现的最大规模的金融危机正促使共和党财政部长和美联储主席大举采用政府干涉经济的措施，希望以此使经济免遭如同大萧条时期一样的巨大破坏。三月，美联储蓄废除了一个近半世纪的规定，即美联储只借钱给那些政府确认交付过保证金的银行。美联储宣称目前的情况是极不寻常而又十分紧急的，所以采纳很少使用的法规是必需的。美联储曾拨款拯救投资银行Bear Stearns ，美联储动用290亿纳税者资金已经向摩根大通公司提供资助，帮助后者收购贝尔斯登公司。这在当时看起来是一笔大买卖，但是在过去的两星期，美国政府开始采用大量反自由主义经济学的政策，成为私有企业的持有者。

－美国政府接管两大住房抵押贷款融资机构房利美和房地美公司（“两房”），用拿纳税人的钱注入抵押市场是其能够继续正常运作。

－美国政府对美国最大的保险巨头AIG实施拯救计划，强力注资AIG并解雇了AIG的首席执行官，对AIG资产进行拆分出售。

－在货币市场基金中实际将政府存款保险扩大到3.4万亿美元。

－暂时禁止799个主要的金融股票被卖空以防这些买空投机者从不断下跌的股价里获益。

－允许和鼓励美国四大投资银行中的第四大投资银行雷曼兄弟破产，美国第三大投资银行美林则被出售。

－要求美国国会到下周务必同意从金融机构冻结纳税人数千亿美元的非流动资产以帮助这些金融机构提高资金存量并能够继续开展借款业务

不到一星期前，美国财政部长鲍尔森在政府的救援计划里指出不会像之前给贝尔斯登救援那样再由政府出面来拯救雷曼兄弟，美国政府允许雷曼这样的投资银行倒闭破产。马萨诸塞州的民主党人、众议院金融服务委员会主席法兰克这周初讽刺说“政府的承诺仅仅持续了一天”。9月15日周一这天之前美国政府刚刚拒绝雷曼的求助，而这天之后美国政府却对AIG施以救援。 Financial Times评论道美国政府将在金融体系中扮演自20世纪30年代“大萧条”(Great Depression)以来和平时期的的最大角色，在许多人眼里，这似乎标志着里根式放松管制时代的终结。

参考[3] 英国BBC的著名财经编辑 Robert Peston：在过去的数年里，很大一部分银行高管一直不负责任的不按照银行的贷款规定办理相关贷款业务。

参考[4] Robert Peston：我们现在看到的是一个失去理性的股市，是一个完全被歇斯底里的热情所推动的股市。