Fated to a big trend slowdown?
Many analysts have predicted that a Chinese economic slowdown is inevitable because the country is approaching the per capita income at which growth in other countries began to decelerate. However, China may escape such a slowdown because of its uneven development. An analysis based on episodes of rapid expansion in four other Asian countries suggests that growth in China’s more developed provinces may slow to 5.5% by the close of the decade. But growth in the country’s less-developed provinces is expected to run at a robust 7.5% pace.
This article highlights that, regardless of the depth of the cyclical downturn, the long term trend is of great concern. The authors provide this graph to illustrate how their ECM of growth in four East Asian economies  can be applied to regions in China to show how prospects might diverge.
Figure 3 from Malkin and Spiegel.
This article is part of a longer debate that has been running – namely whether there is a middle-income trap. On one side is Eichengreen, Park and Shin (2011), who argue for a $17,000 threshold (constant international dollars), (and parenthetically argue that the slowdown is more likely for countries that maintain undervalued exchange rates).
One critique argues that one wants to see at what ratio to the leader’s income per capita a country’s growth slows down; implicitly, this cuts out of the sample countries that remain mired in a low income trap. Andrew Batson at Gavekal notes that deceleration in growth (where the criterion for deceleration is not cited) can occur at quite wide-ranging ratios: 20% for Thailand, 50% for South Korea and 80% for Finland. This is shown in Figure 4 from Batson (2011).
Figure 4 from Andrew Batson, “Is China heading for the middle-income trap?” China Economic Insights (GaveKal Dragonomics, September 6, 2011). [Not online]. Dotted lines pertain to ratios at which growth slowed.
Which one is a more appropriate approach? In my mind, it depends on what the question is. If one wants a characterization of income per capita for all countries, the Eichengreen et al. methodology is more appropriate. If one wishes to condition on the prospect of convergence (keeping in mind convergence is quite rare), then one might want to use the latter approach.
For more on China, see here.