RBA Annual Conference – 2016 Discussion

1. Huw McKay

It is an honour to have been asked to participate in this important conference. It is, however, a little intimidating having to follow such an entertaining and engaging speaker as Professor Wing Thye Woo.

The question he tackled for us – China's prospects for avoiding the middle-income trap (MYT) – is an immensely practical one in which we all have a personal as well as an intellectual interest. Personally, I have been thinking about this question from a number of different perspectives over the course of my career – as an academic, as a banker, as a policymaker and now, as an employee of the world's largest diversified mining company, BHP Billiton. The following thoughts owe something to each of those perspectives.

To begin, my first observation with respect to the MYT is that it is a question in which data constraints loom large. There are two concerns here: the length of the time series and the representativeness of the samples. They are related to each other. As it is a difficult matter to assemble rich economic datasets representing the middle-income phase of economic development in the first- and second-generation industrialisers, researchers tend to fall back on more recent history where such data is more readily available. This inevitably restricts the sample size. This in turn opens up questions regarding sample representativeness.

Limitations related to the sample are often then compounded by the fact that some researchers respond to this challenge by resorting to cross-sectional point-in-time analysis to study MYT issues. Such analyses are very sensitive to sample choice and tend to highlight the reality that at any given time, the middle-income area is lightly inhabited. Successful economies transit the phase rapidly. Few economies enter into the discussion to begin with. Some have entered into the discussion recently by growing into the middle-income region; others have entered into the discussion by falling from grace. These very different paths towards middle-income – recent success and recent failure – are not differentiated in such a framework. In sum, middle-income economies are by definition atypical, being scarce, and are therefore less amenable to empirical rigour than their low- and high-income counterparts.

The most popular and defensible approach from this point is to prosecute a case for why a small sample can be representative and to lean principally on qualitative analysis of the sample constituents, which is what Wing has done. With the east Asian development model on his side, this approach is not particularly controversial. What is more controversial is the way that the various ‘lessons’ of the east Asian path are projected onto China and what examples from outside the region, such as those from Latin America, are seen as legitimate inputs to the discussion.

Here we enter into the zone where political economy, geoeconomics, geopolitics and mainstream economics intersect – somewhat uncomfortably. This is very much the case in one important element of this debate – which Wing left undeveloped in his remarks in favour of the ‘geos’ – that is, the distribution of income. The income inequality variable has been highlighted as China's most notable divergence from the successful path of Japan and the newly industrialised economies (Feng and Yao 2014). A pessimistic interpretation of this observation is that the Chinese leadership's allowance for the fact that some people might get rich before others was a sound strategy for achieving middle-income status, but not for exceeding it. That position is consistent with the Olsonian view that the rise of distributional coalitions – powerful rent seekers with material macroeconomic reach – is inimical to achieving durable growth (Olson 1982). A more positive approach would be to argue that Kuznets (1955) may have been right after all, and that the distribution of income is a function of the development phase and is not a predictor of future outcomes. A side issue here, of course, is that Kuznets' original hypothesis, put forward in the 1950s, was only really questioned when Japan, and later South Korea and Taiwan, broke the mould. A second side issue is that the original Kuznets curve is an egregious example of the difficulties associated with generalising from a limited sample (McKay 2008).

Returning to the issue of inequality and the MYT, Wing was co-author, along with Xiaolu Wang, of a terrific study of income inequality in China (Wang and Woo 2011). They utilised an independent survey of households to underpin their study and vastly improved our knowledge of this issue. So he has far more direct credentials in this field than I do. Where I may have something to contribute is in the practical application of the best research on MYT questions into forming a coherent view on the likely future growth rate and structure of the Chinese economy, and what it will mean for commodity and financial markets. Here the question is not just related to the probability of China entering the MYT or not – the path towards the MYT is also extremely pertinent.

Let me say very clearly at this point that BHP Billiton does not anticipate that China will fall into the MYT. Our view is that China will gradually converge on frontier per capita growth rates as it continues to benefit from the absorption of technology, which will allow it to ascend the value-added chain as its comparative advantage in labour-intensive pursuits fades and its factor markets are progressively liberalised.

Having put those cards on the table, let me return to hypotheticals. There appear to be two basic theses with respect to China and the MYT. The first is that Chinese economic growth falls in a stepwise fashion from its current level to something resembling that of the frontier economy, the United States, similar to the dynamics highlighted in Barry Eichengreen's empirical work on ‘growth slowdowns’ (for instance, Eichengreen, Park and Shin (2013)). The second thesis is that China may follow a more disruptive path, whereby the economy falls into recession, experiencing the proverbial ‘hard landing’. Following this setback, it is no longer able to consistently grow materially faster than the frontier and is accordingly ‘trapped’.

It is very important to emphasise that these are not the only options – they are just the most talked about. We must also recognise that there is a possibility that a recession could occur, but the economy may rebound successfully and ultimately defeat the MYT. In this regard, it is instructive to remember that South Korea's prospects for achieving high-income status were written off twice in the post-WWII era; once in the aftermath of civil war and once after the Asian crisis. And going a little further back, the incidence of 13 canonical financial crises in the United States in the century prior to the establishment of the Federal Reserve System did not prevent that economy from rising to a position of global primacy. All of which is to say that even if a recession were to occur in China before, say, the end of the current decade, that would not necessarily resolve the MYT debate.

And finally, and perhaps somewhat controversially, I note that while we speculate a great deal about the MYT, even the so-called optimists, who do not think that the MYT is a material risk, seem reluctant to project China's ultimate position in terms of relative living standards beyond, say, two-thirds of US levels. Is this a consistent posture? Are the implied constraints that prevent these forecasters from predicting that China will fully converge to US income per capita levels in the long run not also constraints that would impinge upon the successful navigation of the MYT? Decades from now, will we be wringing our hands about a two-thirds or three-quarters income trap, much as we suffer angst about the MYT today? Is there something of a Sinitic glass ceiling for GDP per capita? Some in the long-term projection game would have it so.

While a very wise economist once quipped that in the long run we are all dead, tier one mining assets may have a reserve life that overlaps multiple human generations. Ergo, the hypothetical questions I have run through above are not abstract to BHP Billiton: they are very real problems on which we need to form views. In this regard it is comforting to be able to stand on the shoulders of such scholars as Wing Thye Woo, who tackle the big issues without fear or favour.


Eichengreen B, D Park and K Shin (2013), ‘Growth Slowdowns Redux: New Evidence on the Middle-Income Trap’, NBER Working Paper No 18673.

Feng Y and Y Yao (2014), ‘The Middle-Income Trap and China's Growth Prospects’, in L Song, R Garnaut and F Cai (eds), Deepening Reform for China's Long-Term Growth and Development, China Update, ANU E Press, Canberra, pp 133–158.

Kuznets S (1955), ‘Economic Growth and Income Inequality’, The American Economic Review, 45(1), pp 1–28.

McKay H (2008), ‘Metal Intensity in Comparative Historical Perspective: China, North Asia, the United States & the Kuznets Curve’, The Australian National University, Global Dynamic Systems Centre Working Paper No 006.

Olson M (1982), The Rise and Decline of Nations: Economic Growth, Stagflation and Social Rigidities, Yale University Press, New Haven.

Wang X and WT Woo (2011), ‘The Size and Distribution of Hidden Household Income in China’, Asian Economic Papers, 10(1), pp 1–26.

2. General Discussion

Discussion initially focused on the methodology used in the paper. One participant noted that, while the analytical approach of the paper was restricted by the small sample of ‘middle-income’ economies available, they did agree with Wing Thye Woo that insights could be gained by comparing China's development to the experiences of other economies, particularly Malaysia and the former Soviet Union.

Another participant sought clarification about the definition of the ‘middle-income trap’. They believed the notion that countries in different regions can settle at different fractions of the US standard of living, with no strong economic forces pushing them to the frontier, was quite a diffuse definition of the middle-income trap. A third participant pointed out that all measures of the middle-income trap are imperfect. However, they noted that having an analytical measure is important when making judgements about development through time and across economies, and that the measure used in the paper is sufficient for this.

Dr Woo reiterated that defining high- and middle-income groups should be based on a ratio rather than a particular income level. He added that, although there was some confusion among participants, there are three distinct groups of economies along the distribution. These groups of economies have been fairly stable over time and do share common traits. Dr Woo explained that middle-income economies are those clustered around the 30 per cent of US living standards and that there seems to be something about this level that many economies cannot move beyond.

The discussion then turned to the role of political reform in overcoming the middle-income trap in China. One participant pointed out that political commitment to reform is essential for moving beyond the middle-income trap. They emphasised the issues in China around industrial structural change and the devolution of power away from the centre to a more private sector-driven economy. They also highlighted that financial reform is a critical element in every country's effort to move through the middle-income trap and that this is an area where China has actually done quite well.

Related to this, a participant asked Dr Woo whether he thought initiatives proposed by the Chinese authorities, such as the ‘One Belt, One Road’ initiative or the China-led Asian Infrastructure Investment Bank (AIIB), were possible solutions for the issue of excess capacity of resources Dr Woo described and potential circuit-breakers for some of the dynamics around the middle-income trap. Another participant drew out Dr Woo's point on political consistency being especially important for a country moving into a new stage of development. The participant pointed out that political consistency would be quite difficult for policymakers to maintain if they were to implement the comprehensive set of policies presented by Dr Woo. They stated that while consistency is important, if policy advice is provided to leaders, the focus should be on a central policy issue. The participant believed the key focus should be on enhancing China's ability to upgrade its industry from a low-cost base to a high value-added, high-technology and competitive industry.

To address questions relating to the Chinese government initiatives, Dr Woo agreed that the AIIB would provide a channel through which China could export excess iron and cement, and the construction of an express railway between Beijing and Moscow also has potential to absorb some of the excess capacity. However, Dr Woo stated that it was more important that China expand its trade and build relationships within its geographical neighbourhood. Dr Woo highlighted that, in addition to geopolitical tensions, China's excess production capacity is so high that it tests China's trade relations with the rest of the world.

One participant noted that China achieved strong economic growth over the past decade, despite the many challenges it faced. The participant asked Dr Woo whether, given this evidence, a perfect system is really needed to continue economic growth. Dr Woo agreed that a perfect system is not needed to advance and get beyond the middle-income trap; otherwise no country would have ever moved past it. He highlighted that it is very important that China upgrade its industry by implementing structural reforms and that these reforms are communicated to the Chinese public in a way that minimises any social dissatisfaction or political backlash.