The Financial Times carries a comment by the author of this blog on a column by Martin Wolf, its chief economics commentator, on the development of consumption in China.
Martin Wolf has a justified reputation as probably the world's most influential economics columnist. However, in this case, his argument confuses the issue of the low percentage of consumption in China's GDP with the rate of growth of China's consumption and its consequences for the population's living standards - China has the world's fastest growth rate of consumption and it is this high rate of growth, not the percentage of consumption in GDP, that determines the most rapid development of living standards and total consumption.
The practical economic significance of this is that in every country, including China, the growth rate of consumption is closely correlated with the growth rate of GDP - as discussed in the article below. In turn GDP growth is correlated with both the quantity and efficiency of investment. Increasing the percentage of consumption in China's GDP via a lower level of investment, other things remaining equal, would therefore lead to lower GDP growth rate, and lower absolute growth of consumption and living standards, than maintaining a higher investment percentage.
China’s present investment level would be unviable only if either it led to falling profits, which is contradicted by the data, or its efficiency of investment from the point of view of GDP growth, that is the ratio of fixed investment to GDP growth rate. was abnormally high – in fact it is almost exactly the same as India’s and around half that of the US, i.e. China's efficiency of investment, from the point of view of GDP growth, is the same as India and almost twice that of the US.
The issues of the relation of investment and growth, and of the relative efficiency of China's investment compared to other countries, have been dealt with several times here. The article below, however, deals with the distinction between the percentage of consumption in GDP and the growth rate and level of consumption. It originally appeared, in Chinese, in the journal China Finance and was therefore aimed at a non-specialist economist audience. Readers are referred to the Financial Times for a more compressed version of the arguments. The original title of the article was 'What should be targeted in consumption?' Its main conclusion is that the target of economic growth should be 'the maximum sustainable rate of development of consumption'. A few purely contemporary issues have been eliminated so as not to clutter the main argument.
* * *
What should be targeted in consumption?
This article analyses issues in the development of consumption in China – notably which macro-economic measure should be targeted in this. The significance of this issue is at least twofold.
- Strategically, sustainable consumption is the goal of economic activity – investment is only the means to achieve this.
- More immediately, in the situation China’s economy found itself in at the time of the outbreak of the financial crisis, with a net trade surplus amounting to almost nine percent of GDP, reorienting the economy to domestic demand evidently meant raising the percentage of consumption in GDP.
However this latter immediate issue does not settle the question of by which percentage consumption should be increased as a proportion of GDP, what should simultaneously happen to investment, and what should be the framework for measuring the strategic development of this matter?
The significance of this question is made clear by the fact that at least part of the discussion of this issue confuses the rate of growth of consumption with the percentage of consumption in GDP – the two are very different. The implications of this distinction and different formulations on the issue of consumption are also illustrated below in regard to key current economic topics - inflation, house prices, and wages.
In developing a precise response on many of these issues self-evidently the most important answers can only be defined in China. However some questions are subjects of general economic theory. It is regarding some of these latter issues that there appear to be questions which may benefit from international discussion.
The goals of economic policy
The target set to guide economic development cannot be defined independently of a specific situation and goals. To take an extreme example, during war all other goals (long term productive investment, individual consumption etc) must be subordinated to maximum development of military potential. For a prolonged period after the commencement of China’s reform and opening up process in 1978 the goal was set of the most rapid possible GDP development. This corresponded to the priority of the development of the productive forces and overcoming China’s position as a low income country.
Currently in China there is widespread agreement that a goal of the maximum possible tempo of GDP increase is now not appropriate for well know reasons. If the priority is the maximum rate of GDP growth then environmental considerations are by definition subordinate – which has cumulatively led to negative environmental degradation, and similarly social and environmental inequalities are secondary.
What strategic goal therefore is appropriate in regard to consumption? The most accurate formulation of this is ‘the maximum sustainable rate of development of consumption.’ The reasons for this are set out.
Confusion of the percentage of consumption in GDP and the rate of growth of consumption
The conceptual error in confusing the rate of growth of consumption with the percentage of consumption in GDP, and why it is rate of growth and not the percentage share that is crucial, may be illustrated clearly by considering the factual development of consumption in the major economies. This is shown in Table 1 - which notes the rates of growth of consumption in the most rapidly growing developed and developing economies.
It is well known that China has a high share of fixed investment in GDP and a relatively low percentage of consumption in comparative international terms. However it may be seen from Table 1 that China has had by far the highest rate of growth of consumption of any country in the world since the commencement of its reform process.
China’s rate of growth of total consumption, that is both household consumption and state spending on items such as education, health etc, has risen at 8.7% a year over the reform period compared to 6.1% for Malaysia, 5.9% for Indonesia, and 5.6% for South Korea - the next best performing major economies in regard to consumption growth. China’s 8.7% annual rate of growth of total consumption evidently far exceeds the 3.0% for the US, 2.6% for the UK, 2.5% for Japan, 1.6% for Germany etc.
China has also had by far the fastest annual rate of growth of household consumption – 8.6% compared to Malaysia’s 6.8%, Indonesia’s 6.0%, South Korea’s 5.6% etc. China’s rate of growth of household consumption is evidently far higher than the US’s 3.5%, the UK’s 3.0%, Germany’s 1.6% etc.
Table 1
Correlation of rate of growth of consumption and rate of growth of consumption
The reason China has one of the lowest percentages of consumption in GDP, but the highest rate of growth of consumption is, of course, because it has the highest rate of GDP growth.
Table 1 evidently confirms that all countries with the highest rate of growth of consumption – China, Malaysia, Indonesia, the Republic of Korea, Hong Kong, Thailand, and India – have the highest rates of growth of GDP. Countries with low growth rates of GDP – the US, UK, Japan, France, Germany etc – also have low rates of growth of consumption, although they have higher shares of consumption in GDP than China. There are no countries with a high rate of growth of consumption and a low rate of growth of GDP no matter what is their share of consumption in GDP.
This illustrates clearly the decisive point that what is crucial in the economic development of consumption is its rate of growth, in which the determinant factor is the rate of GDP growth, and not the percentage of consumption in GDP.
It is evidently of no interest to a person what is the percentage share of consumption in GDP. They cannot eat a percentage or go on holiday on a percentage. What is important is their actual consumption and their rate of growth of consumption. Similarly, for production the rate of growth of the market for consumption goods is determined by the rate of growth of the increase of consumption, not by a percentage of consumption in GDP. It is therefore the rate of growth of consumption, not its percentage share in GDP, that must be targeted if the goal is to develop consumption. Indeed as will be seen it is China’s high level of investment which creates the basis for its rapid economic growth and it is this in turn which permits its high rate of growth of consumption.
Division of labour and economic growth
The actual factual quantitative interrelation between GDP growth, investment and consumption can be illustrated clearly by looking at the results of modern econometric research on the causes of economic growth – particularly economic growth in developed economies. That this is the crucial area to concentrate on is already established by the fact shown above that the most important factor in the growth of consumption is the rate of growth of GDP.
Analysing first the US economy, the sources of its GDP growth are shown in Table 2. This lists the normal inputs into GDP utilised in growth accounting - capital, labour, and total factor productivity – plus a measure of the increase in ‘intermediate inputs’ in production, that is of non-finished products that serve as outputs of one industry into another. The reason for including this last element is that modern econometric research shows that the rise of intermediate inputs grows more rapidly both than GDP or the factors of production, labour and capital, or total factor productivity.
It is easy to give an economic interpretation of the statistical finding that the rate of growth of intermediate inputs is more rapid than the rate of growth of GDP – it also gives direct econometric vindication of China’s ‘opening up policy’ and the Chinese government’s firm stance against protectionism in the present international financial crisis.
The fact that intermediate inputs grow more rapidly than GDP, in the case of the US at 115% of the rate of GDP growth, shows that that the division of labour within the economy is increasing. It measures in modern econometrics the famous opening sentence of Adam Smith's The Wealth of Nations: 'The greatest improvement in the productive powers of labour, and the greater part of the skill, dexterity and judgement with which it is anywhere direct, or applied, seem to have been the effects of the division of labour.'
In a modern economy this division of labour is international in scope. Comprehensive statistical data does not exist for the growth of intermediate inputs in production on an international scale in the way that it exists within the US economy. Nevertheless the fact that the fundamental trends of globalisation of a rising share of trade in GDP, and a rising ratio of foreign investment to GDP, plus the studies of particular industries, are powerful evidence of an increasing division of labour on an international scale – and there is in any case no reason to believe that the processes creating the clearly measurable increasing division of labour on a national scale should not operate internationally. Attempts to ‘close’ an economy therefore necessarily cut it off from the widest, that is a necessarily international, division of labour. Modern Western econometric analysis therefore is entirely in accord with the conclusions of China in its ‘opening up’ process.
Investment and economic growth
After increasing division of labour it is clear from Table 2 that investment, that is increase in fixed capital, is the primary element in US economic growth. Fixed investment accounted for around half US GDP growth and shows no tendency to decline.
Considering the other factors of production the significant shift is that the increase in total factor productivity has replaced increase in labour inputs as the second most important quantitative factor in US economic growth after fixed investment. To the degree that total factor productivity is traditionally taken to indicate technology, this means that over time the US has shifted from a ‘capital plus labour’ path of GDP growth to a ‘capital plus technology’ path. This development evidently creates problems for the US economy in generating jobs during growth, a problem which is also an issue in China, but it clearly underlines the constant emphasis placed by China’s economic policy on technological development.
Table 2
Determinants of economic growth in the G7 economies
Turning from the US to the G7 economies as a whole the factors in growth for these are set out in Table 3. Regrettably quantitative data for the growth of intermediate production does not exist at an international level, but the ranking of the order of magnitude of factors of production in GDP growth for the G7 countries is evident.
As with the US growth in fixed investment is the most important factor in G7 growth. The mean contribution of fixed investment to GDP growth for the G7 is 60.3% and the median contribution is 53.6%. Therefore, as with the US, half or slightly above half of GDP growth is accounted for by fixed investment growth.
The mean contribution of increase in labour hours is 11.3% of G7 GDP growth and the median contribution 13.9%. The mean contribution of total factor productivity growth is 28.5% and the median contribution 31.4%.
Therefore for the G7, as with the US, after the increase in the division of labour the rank order of contributions to GDP growth is first fixed investment, second total factor productivity (technology), and third labour.
Table 3
Sources of growth in developed and developing economies
It is illuminating to compare the determinants of GDP growth for developed economies with those in developing economies. Table 4 therefore shows the mean and median contributions of factors to economic growth for 22 developed and 87 developing economies for which the necessary data is available.
The developing economies are far more dependent on the growth of labour hours than the developed economies. Taking means for the groups, the contribution of inputs of labour hours to GDP growth is only 23.1% for the developed economies group compared to 38.3% in the developing economies. Taking medians the figure is 23.8% for the developed economies compared to 40.6% for developing economies.
In contrast there is little difference between the developed economies and developing economies in terms of the contribution of total factor productivity to growth – the mean contribution of total factor productivity to growth is 24.0% in developed economies compared to 25.4% in developing economies, and the median figures are 26.0% in the developed economies compared to 25.0% in the developing economies.
However there is a considerable gap between the developed and developing economies in terms of capital inputs. The mean contribution of capital inputs to GDP growth in the developed economies is 52.9% compared to 36.3% in the developing economies, and the median contribution in developed economies is 50.6% compared to 39.3% in developing economies.
In summary there is little difference between developed and developing economies in the contribution to growth of increases in total factor productivity, but developing economies have a more ‘labour intensive’ path of development while developed economies have a more ‘capital intensive’ one. The implications of this for China are evident. The more its economy comes to be a developed one the greater will be the contribution of investment to economic growth.
Table 4
Consumption, supply and demand
It may be noted that consumption is not present in the factors for GDP growth. The reason for this is evident. Consumption by definition does not increase supply, that is growth – if it did it would be investment not consumption.
Consumption instead increases demand. Demand may, but it also may not, stimulate an increase in supply – this depends not only on increased demand from consumption but on a series of other factors including whether extra production is profitable, whether there is unused capacity etc.
For example, if there is not unused economic capacity an increase in demand, via an increase in purchasing power for consumption without the ability to increase output, simply produces inflation and not growth – a fact highly relevant to present economic discussions as noted below. Demand, of which consumption is a component, therefore by definition cannot directly produce economic growth, it can only do so indirectly via effect on supply.
The relation of consumption, growth and investment
The factual quantitative relation between consumption, growth and investment is therefore clear. If, as it should be, the aim of economic development is to develop the rate of growth of consumption then quantitatively:
- The rate of growth of consumption is itself primarily dependent on the rate of growth of GDP.
- The rate of growth of GDP is most importantly determined by the increasing division of labour – including participation in the international division of labour.
- After growth of division of labour the rate of growth of GDP is primarily dependent on the rate of growth of investment.
- Therefore, the growth rate of consumption is itself dependent on the rate of growth of investment.
- A developed economy is more dependent on investment for economic growth than a developing economy.
The error of confusing the rate of growth of consumption with the percentage of consumption in GDP is therefore clear from the point of view of developing consumption itself. If the percentage of consumption in GDP is raised and this leads to a reduction in the rate of GDP growth, and other things being equal it will lead to a reduction in the rate of growth, then the rate of growth of consumption will be lower. There is, however, self-evidently no point in saying to a person in any country ‘it is true that you do not have more and higher quality food, you do not have a car, you do not have as many vacations as was possible because GDP growth has slowed but you should be content because the percentage of consumption in GDP is higher.’ A person will answer ‘I don’t care about the abstract issue of the percentage of consumption in GDP I want more and better food, a car, and better vacations.’ That illustrates why the target must be the rate of growth of consumption and not the percentage of consumption in GDP.
Why the target should not be ‘the maximum rate of growth of consumption’
Do the above considerations mean, therefore, that the economic target which should be set is the ‘maximum rate of growth of consumption’ rather than the ‘maximum rate of growth of GDP’? No, that would not be a correct target because it does not take sustainability into account. Undesirable consequences would therefore flow from setting the target as the ‘maximum rate of growth of consumption’. Consumption growth, as with GDP growth, must be sustainable.
What is considered, from a national point of view, as ‘sustainable’ is really for China to decide. But some issues other than development of consumption are clearly generally agreed.
- In all countries the rate of growth of personal consumption is limited by defence expenditures.
- Environmental protection is a priority in defining sustainability in China, which will require expenditure and restrictions which limits the maximum rate of growth of consumption.
- There appears to be a growing consensus that the development of inequality in China has become excessive and therefore social sustainability requires measures to lower the level of at least certain forms of social and economic inequality.
The shift from foreign to domestic demand
In order to illustrate the practical, as well as theoretical, economic importance of some of the above distinctions their implications for certain key immediate issues of the economic situation will be considered.
The first is the shift from international to domestic demand. China entered the international financial crisis with a trade surplus that amounted, at official exchange rates, to nearly nine per cent of GDP - with a roughly equivalent balance of payments surplus. This was clearly too high and a shift into domestic demand was in any case highly desirable even if it had not been imposed by the fall in exports due to the financial crisis. If the wholly reasonable target were set of eliminating the trade surplus entirely this means that approximately nine per cent of GDP could be shifted into domestic demand – i.e. into domestic investment, domestic consumption or both.
Naturally a shift of 9% of GDP cannot be achieved in one year and in 2009 about 4-5% of GDP was shifted from the trade surplus into domestic demand - allowing a simultaneous increase in both the percentage of consumption and the percentage of investment in GDP. The fact that the percentage of both consumption (demand) and investment (demand and supply) was increased meant that both the demand and the supply sides of the economy were strengthened - a desirable consequence. The fact that the supply side of the economy, as well as consumption, was reinforced will, for the reasons outlined above, strengthen the most rapid sustainable development of consumption.
What, however, would have been the consequence if this entire shift had taken place into consumption, that is if a target of ‘maximising the percentage of GDP devoted to consumption’ rather than ‘maximising the sustainable rate of development of consumption’ had been set? It would have been that the demand side of the economy would have been strongly stimulated but there would have been no domestic increase in supply – as by definition consumption does not add to supply. The result would have been even stronger inflationary pressures than actually existed. In short it would have worsened the situation in the economy to have shifted all resources released by the decline in the trade surplus into consumption.
Equally, however, a shift of 4-5% of GDP, let alone all 9% of GDP accounted for by the trade surplus, solely into investment would have evidently been excessive – there are no international examples of such a rapid build up of investment being deployed effectively. The correct choice was therefore made to expand domestic demand both by increasing domestic consumption and domestic investment – this should help sustain a strategic target of the most rapid possible sustainable increase in consumption.
Conclusion
It is evident from the issues above why the decision changing the goal of economic development in China from ‘the maximum rate of growth of GDP’ was correct from the point of view of economic theory. It is also clear that shifting the goal into consumption is right. However in economics it is not sufficient only to have good intentions. It is necessary to formulate the target accurately if the maximum possible living standard of the population, which depends on consumption, is to be achieved.
The formulation of the precisely required forms of sustainability is evidently an issue for China itself. However it appears that the formulation of at least two issues regarding the development of consumption can be aided by discussion – including by international discussions. These are that:
- The development of consumption must be formulated in terms of the rate of growth of consumption and not the percentage of consumption in GDP.
- A goal must be set in terms of the maximum sustainable rate of development of consumption and not simply the maximum rate of development of consumption.
