Wednesday 06 Oct 2021 | 20:09 | SYDNEY
Wednesday 06 Oct 2021 | 20:09 | SYDNEY

Yes, China really is catching up


Mark Thirlwell

This post is part of the US-China: Measuring decline and rise debate thread. To read other posts in this debate, click here.

15 February 2012 08:25

This post is part of the US-China: Measuring decline and rise debate thread. To read other posts in this debate, click here.

I'm sympathetic to the distinction Michael Beckley raises between GDP and GDP per capita in his post on defining decline; it's a point I also focused on in my original post. Each gives us different readings about national capabilities.

But the other question I raised concerned comparisons over time. Specifically, I was surprised by Beckley's contention that the economic gap between the US and China had grown since 1991, and I plotted some GDP and GDP per capita data to suggest that the gap looked like it had shrunk. Beckley disagrees and the source of this difference is that, while I focus on ratios, including the ratio of GDP per capita, Beckley looks at the absolute difference in GDP per capita between the two economies.

It certainly makes sense to look at the absolute difference when comparing the two economies at a given point in time. But absolute differences do not do a good job of capturing changes over time. That's why we should normalise the data when making these kinds of comparisons – for example by constructing ratios or indices.

To see why, take the following (deliberately simplistic and extreme) example. Suppose that, at the start of our comparison period, GDP per capita in the US was $1000 while GDP per capita in China was just $1. Then assume that by the end of our comparison period US GDP per capita had risen to $1.001 million while Chinese GDP per capita had risen to $1 million. So China has grown much faster than the US over this period, but the US remains the wealthier economy: 

By the end of this simplistic example, has China shown signs of catching up with the US or not?

If we look at absolute values, we would have to say 'no'. In fact, the gap between the two countries has increased, rising from $999 at the start to $1000 at the end. And yet the message that China has gone backwards relative to the US is incredibly counter-intuitive.

So instead, let's look at the ratios. On this metric, while Chinese GDP per capita was just 0.1% of US levels at the start, it has climbed to 99.9% of US levels by the end. This is an extreme (and unlikely) example of catch-up growth that is nevertheless consistent with the US remaining wealthier, and with the absolute gap between the two measures of GDP per capita having increased.

But doesn't using ratios give an exaggerated picture of China's catch-up? (Beckley said in his post that, by my definition, 'the US has been in decline to China since 1968...') Well, not really. The chart below still suggests that the catch-up story only gets going after the start of the reform period*:

I suspect part of the difference in emphasis here is due to conflating the ideas of catch-up and decline. 

From my perspective, the former need not imply the latter. In other words, the Chinese economy can converge on the US economy without the latter declining, since the catch-up growth story is perfectly compatible with a world in which both economies can keep growing and in which the US can also keep getting richer in absolute terms (see my simplistic example above). Note that my original post does not mention US decline once.

The security debate sometimes seems to treat Chinese catch-up as necessarily equivalent to US decline. Perhaps this reflects an important difference between positive-sum-style economic thinking and zero-sum-style strategic thinking, a difference in approach I have highlighted elsewhere.

* In order to graph both series back to 1950, I have used data constructed by the economic historian Angus Maddison. In my original post I used data from the IMF's World Economic Outlook database, so I have included that too, for comparison. I discussed the big difference between the Maddison and IMF data on GDP per capita (and GDP) as measured using PPP rates in this post.