Montag, 1. Juni 2009

Is China Investing Too Much?

Just about every outside observer seems to agree that China is investing too much.

And the numbers sure look that way:

- China's gross investments make up close to 50 % of GDP.

(Edit: Checked the official national accounts data, and the exact number was "only" 42-43 % in 2004-07)

For comparison:

- Germany records around 20 %

- The US barely manages 15 %.

Already back in the 90s, Krugman famously argued that Asia's "tiger economies" were using more and more capital to achieve their high growth, whereas total factor productivity remained pretty constant. Based on this finding, he argued that the Asian "growth miracle" wasn't particularly miraculous after all.

However, high amounts of capital investment don't necessarily imply inefficient production or overinvestment. This is only the case if the return on the capital employed is not good. Countries can have a different economic structure: For instance, the US has comparatively little capital-intensive manufacturing, whereas Germany has lots of it. This probably explains why Germany has a higher investment rate than the US, even though the US economy has been growing faster than Germany's.

However, the difference between 15 % in the US and 20 % in Germany is dwarfed by China's 50 %. Can such a gargantuan investment rate possibly be adequate and sustainable?

Well, there might be a reason to argue that indeed it can be:

- Germany's economy is basically stagnating (average real growth over the last ten years has been barely more than 1 % p.a.)

- China's economy has been growing at roughly 10 % p.a.

When an economy grows fast, the capital stock also needs to grow fast. This means that new investment as a percentage of GDP must be higher in high-growth countries, otherwise the capital stock will shrink as a percentage of GDP.

Let's illustrate this with numbers:

- Assume that on average, investments have a useful life of 15 years

- In a stable economy such as Germany, a 20 % investment rate implies a total "equilibrium capital stock" of 300 % of GDP (20 % * 15 years).

- Assume China's economy is structurally the same as Germany's, i.e. also has an equilibrium capital stock of 300 % of GDP.

- If China grows at 10 % every year, it needs to grow its equilibrium capital stock at 10 % as well. As the capital stock is assumed to be 300 % of GDP, 10 % growth implies that 30 % of GDP need to be newly invested every year on top of replacement investments.

- China's replacement investments will be lower than Germany's 20 %, as the economy has grown fast, so what used to be 20 % of GDP is far less by the time the investments have reached the end of their useful life. So China's replacement investments are probably more like 10 % of current GDP.

=> Based on the assumptions above, China would have equilibrium investments of 40 % of GDP as compared to Germany's 20 %. The difference is needed to keep the capital stock constant in % of GDP while the economy is growing at 10 %.

(If the average useful life of investments is longer than 15 years, equilibrium investments will be even higher than 40 %. If the useful life is shorter, they will be below 40 %.)

So China's high investment rate doesn't necessarily mean that there is a huge degree of overinvestment and lots of capital is wasted: A high investment rate is perfectly in line with a capital-intensive economy (similar to Germany's) which needs to expand its capital stock to keep up with high overall growth rates.

(Note: I do not intend to make the claim that China's investment binge is fully appropriate. I agree with most other observers that China seems to be overdoing things investment-wise. But based on the thoughts outlined above, I do believe that the extent of the overinvestment is probably less than commonly suspected.)


  1. What's India's investment rate?


  2. Good question.

    According to this link:

    India's investment rate was 39 % in the fiscal year ending 3/08.

    Considering that India is commonly perceived as an economy with low capital-intensity, and also considering that India's growth is a bit lower than China's, I'd say that the percentage is surprisingly high. Lower than China's, but not worlds apart.

  3. after 15 years of coninous growth,the GDP will be 4.17 times bigger,than in the first year.
    It mean that the investment will be 12.52 times the year one GDP.

    In year one,the amortisation was 0.2,aka 20% of the GDP.
    It mean that from the year 15 GDP they have to spend the 0.047, aka 4.7% of the GDP for replesnihment(of the year 15 GDP).
    So,if the investment to the growth 30%,the investment into the replenishment 4.7%.So,the natural investment % have to be 34.7% in your model.

    China have 10% GDP from the export,and the 45% of the GDP comming from the consumption.
    If it is growth by a ballanced way by 10% in a year,by your modell they have to have 34.5% investment,and 65.5% consumption after 15 year of continous growth.

  4. Hi Bomlat,

    yes, you're right.

    There are two reasons why I arbitrarily used a higher number (without explaining it in the post):

    1. An average useful life of 15 years doesn't mean that all assets have a useful life of 15 years. Some will live longer, some shorter. Due to the compounding, this increases the need for replacement as a % of GDP to roughly 5.5 - 6.0 % (depending on exact assumptions).

    2. I assumed that China's past investments are of lower quality than in Western countries, i.e. their useful life is lower even if the investment mix is the same. (To make matters simple, I assumed that in the present, the useful life is identical. If that still isn't the case, then everything gets more complicated, and I didn't want to bother to model it in detail)

    But you're probably right that even taking into account those two aspects, the replacement investments are likely to be below the 10 % I have used in the post, given how fast China's economy has grown.

  5. [i]2. I assumed that China's past investments are of lower quality than in Western countries, i.e. their useful life is lower even if the investment mix is the same. (To make matters simple, I assumed that in the present, the useful life is identical. If that still isn't the case, then everything gets more complicated, and I didn't want to bother to model it in detail)[/i]
    It make the number lower,not higher,becasue it mean thet you had to replace this asset with a good qulaity one during the past 15 year,and now it decreasing the required replacement investment

    but from the other side,the many variable make it hard to make an axact calculation

    I think that 3`% investment could be sustainable.It is just a feeling.

  6. I'm not sure I understand your point: All other things being equal, a low quality asset (defined as an asset with a lower useful life) needs to be replaced faster, therefore the replacement rate (in % of GDP) needs to be higher for the same equilibrium capital stock (in % of GDP).

    But in any case, I suppose it's pointless to try and derive an exact number due to all the many things we don't know. The idea of the post was simply to illustrate that high investment rates can be caused by high economic growth.

  7. I agree with the uncertency and so on,but if we talk about 10% GDP from the export,and 10-20% from the investment which not supportable by the real ecconomy then we have real problem in China.

  8. Maybe. Or maybe the Chinese just love lots of brick and mortar and fancy hardware.

    But I agree: China cannot go on investing at this rate forever, and it cannot go on to have a 10 % trade surplus forever.

  9. Checked the National Accounts data on the Statistics China webpage, and the exact investment rate was 42-43 % in every year from 2004 to 2007, i.e. significantly lower than 50 %, and only slightly higher than in India.

    2008 data is not out yet. But fixed asset investment has grown sharply, so the overall investment rate most probably increased to around 45 %.

  10. Dieser Kommentar wurde vom Autor entfernt.

  11. the above comment is not here.
    That I want to say:
    45% investment,45% consumption,10% export.
    And they producing four times more tuck than the US (heavy duty),so I am sure about the investment is wrongly allocated.