RBA stays bullish on Chinese construction

While the rest of world is paring back its assumptions of the potential for further Chinese urbanisation, the RBA is having none of it. In a paper published yesterday (h/t 3d1k) , the RBA made some guesstimates for the path of ongoing construction and there’s no over-capacity in sight as far out as 2040:

This paper assesses the medium-term prospects for the Chinese urban residential construction boom, using projections of urbanisation rates, building size and construction quality. Chinese urban residential construction is expected to remain at elevated levels for the next couple of decades, but growth will undoubtedly slow.

As the urbanisation process winds down and as improvements in building quality become more incremental, construction is expected to peak in 2017 and fall back below current levels some time around 2030.

This projection is dependent upon a number of assumptions, and we consider the sensitivity of our projections to these assumptions. If urbanisation advances at a slower pace than envisaged, then the peak of residential construction could be close to hand. However, there are upside risks as well. The evolution of floor space per capita is quite uncertain; should this advance at a faster pace than the
relatively conservative view in the baseline scenario, the high-growth phase of Chinese residential construction could continue for several more years.

Fundamentally, the construction boom is the result of extraordinary economic growth and urbanisation, which are intertwined in complex ways (Spence, Annez and Buckley 2008). Other countries, for example Korea, have experienced similar,  or even more rapid, increases in the proportion of residents living in urban areas.

Yet no country can compare to China in terms of the sheer scale of its urbanisation. China’s urbanisation process has other distinct features. The hukou system, which determines individuals’ residency status and welfare entitlements, limits labour mobility to a degree that could be responsible for a large gap in wages between urban and rural areas (Henderson 2009). There is also evidence suggesting that Chinese cities are undersized compared to some optimal level, suggesting scope for further expansion in city sizes (Fujita et al 2004; Au and Henderson 2006). This paper does not take account of these features explicitly in the analysis, but they are likely to have only minor implications relative to the influence of China’s overall growth.

Construction requires steel, which in turn requires iron ore, of which Australia is a significant producer. We estimate that residential construction uses about 14 per cent of China’s crude steel output. More intense use of steel, due to taller buildings and other amenities such as underground car parks, means that steel use by residential construction will grow at a faster rate than the volume of floor space built. Indeed, we project that steel used in residential construction will peak around 2024, at a level that is 30 per cent higher than in 2011.

This paper also touches on the structure of the Chinese economy and its evolution. As is well known, Chinese nominal investment as a share of GDP is high. We estimate that the residential construction share of GDP has risen from 5½ per cent in 2004 to 9 per cent in 2011. The extent of this increase is somewhat higher than suggested by the volume of floor space constructed. Quality improvements and higher materials prices could account for this.

I will leave you to read the wonkish meat of the article but there is one point I’ll make. The analysis is based upon other Asia nation urbanisation paths:

But really, is it a good idea to compare a giant, immensely poor, continental economy with small, coastal, mercantile states? Just askin’.

The assumptions of the paper are also revealing:

Given uncertainty about China’s trend rate of growth, we consider two possibilities: a high-growth and a low-growth alternative.

• First we assume that the Chinese GDP per capita grows at 8 per cent per annum for the next 20 years, rather than around 6 per cent in the baseline. Lin (2011) suggests that China has the potential to do so considering China’s stage of development relative to that of the United States, and the growth performance of Japan, Taiwan and Korea when they were at a similar stage of development.

In this case we project that China’s growth rate will follow the same deceleration after 2032 as the growth rate from World Bank and Development Research Center (World Bank and Development Research Center of the State Council, the People’s Republic of China) (2012). Accordingly, China’s PPP-adjusted GDP per capita reaches almost US$80 000 by 2040 (at 2005 prices), compared to around US$40 000 in the baseline projection.

• In the low-growth alternative we assume that GDP per capita growth is 2.7 percentage points lower than the baseline growth rate. This difference is equivalent to one standard deviation of the distribution of growth rates experienced in China between 2000 and 2009.

In this case, China’s PPPadjusted GDP per capita reaches US$20 000 by 2040 (at 2005 prices). Under the high-growth assumption, floor space construction peaks in 2023, 25 per cent above recent levels, and remains above current levels for all but the last couple of years of the projection period (Figure 15). Under the low growth assumption, construction is effectively at its peak now and will be almost 40 per cent below the baseline by the end of the projection period.

So, at least we got a proviso. 8% GDP for the next 20 years? Maybe. I’m happier with 5.3% though even that sounds pretty aggressive. There is also the assertion that only 14% of steel goes into real estate. ANZ estimates 39%. That’s an almighty discrepancy which I can’t explain.

Full report below.


David Llewellyn-Smith
Latest posts by David Llewellyn-Smith (see all)


  1. “8% GDP for the next 20 years”

    28 years as the report forecast period is now until 2040.

    Realistic assumption?

    • “8% GDP for the next 20 years”

      Even Wen Jiaboa would be horrified by that.

      There should be clauses attached to every taxpayer funded research report to clawback a % of the author’s salary if they get it wrong beyond a certain %.

      I’ll probably never have to pay any tax ever again!!

      • Ummmm …. so 8% growth over 28 years means China’s economy will be almost nine times larger than it is today.

        Peak oil anyone?

        The greatest shortcoming of the human race is our inability to understand the exponential function — Albert Bartlett, engineer.

          • Just astonishing that the RBA sees no problem with 1.3 billion Chinese living a western standard of living with western levels of energy consumption.

            Don’t get me wrong, I want to see the Chinese people raise their standards of living — instead of all the wealth going to property developers and Aussie miners — but if China is 9x the size it is today something will break. The Earth ain’t getting any bigger.

          • Global warming? Oh that’s right, that’s a conspiracy by greedy scientists to get more funding. Silly me, the Chinese can burn all the coal and oil they like. We’ll be fine!

          • what happens to the price of those commodities under all those peak scenarios, you think the Chinese economy will become 9 billion westerners at $1000 oil?

            The fertility rate in China is 1.5 well below replacement, they will get old before they get rich, if anything India is more worrisome if your into the peak Armageddon scenarios

          • Mav

            In my lifetime I’ve seen all sorts of extreme language and ideas become commonplace and accepted.
            I, to my great regret, was a participant being of a generation that indulged itself a fair bit.

            I can assure you it’s best not to use it even figuratively. It’s plain dangerous.

            Sometimes MB does indulge itself in extremism. You’ll see it start and gather momentum even here.

            It scares the hell out of me.

    • that a growth to more than 8 fold !!!! where will they find the oil/energy/material/exports to achieve that ??

      good luck in fantasy land

      • The RBA missed the GFC, missed oz economic forecasts, missed the Chinese credit bubble, missed iron ore price halving and denies that oz has a housing bubble.

        Why should anyone listen to the RBA anymore? They are behaving like copywriters promoting the image of a strong Australia. Glenn Stevens admits as much himself.

  2. Clearly our economic future has been placed in the hands of mad men.

    Batten down the hatches people!

    • The RBA seems to have all the same powers of prediction shown by the FED’s Bernank before the GFC.

    • Oh you awful pessimist Lorax! How Un-Australian.

      Crack open your purse and let the moths out mate. Get a mortgage or three. It’s happy happy times and “a period of unprecedented prosperity .. for several decades .. stretching to 2050*”, don’t ya know.

      (* h/t Ken Henry, Oct 2009)

  3. I don’t want to loose faith in the RBA completely, but they are worrying me. I don’t know how you see into the future given the base situation in China come up with this. Also, given the GDP is more like 4% according to the Chinese economist’s I talk to as well. If they are using PBoC figures when even the premier says they are more like a guide …

    it’s either irresponsible of them or the usual spin until you just can’t anymore.

    I think this is part of a cycle, but the lows I think we’ll get make me believe that the RBA is looking at different data, or they know something everyone else doesn’t.

    • I used to think they knew things we didn’t too … Then when the evidence rolled in I realised they were just wrong 🙂

    • Prepare for them to get wackier by the month as the cognitive dissonance hits full stride, al a Greenspan.

    • a63,
      Maybe Glen thinks now that he is getting such a big salary he has to create the impression that a)he is working for it b) the RBA is top of things and the future looks good.

      It appears to me that the RBA has taken on a marketing mission, which conveniently happens to be pro- kool aid (and pro Govt ). A quid pro quo for the hefty salary rise?

      I still beleive the RBA is smarting from it’s appalling actions in 2007/8 , raising interest rates in the face of then exploding GFC.

  4. I made the point in a discussion on economic models with SoN that there aren’t immutable laws in economics, just models with reasonable or unreasonable assumptions. This paper showcases what boffins would call an ‘epic fail’ assumption.

    As if in response, Andy Xie has a new piece up calling $50 iron ore http://my1510.cn/article.php?id=83900

    • MJV,

      “$50/ton is probably the price to expect in the long term. It is twice the price in the 1990s. Taking into account of the cost increase, this price is sufficient to keep the mining industry profitable.”

      It’s going to kill the small producers, and I can’t see explorers having a business case at that price. Over the very long term it would have to rise else who’d be bothered in putting capital into that sector?

      Also, would a $20 producer be interested or re-direct capital? I don’t know, but $50 to me look too low.

      • I have no idea if it’s too low. I would have thought so, but I haven’t seen enough data to take an informed position on it.

        He makes an important point though that production capacity is being increased massively despite there being very little prospect for further dramatic increases in iron ore demand. China’s current rates of steel consumption are the result of distorting and unsustainable policies, so they will fall back one way or another; either by adjusted policies sooner or a crash later. Critically, as Xie says, China’s annual steel consumption doesn’t need to grow for its per capita stock of steel to keep growing rapidly. Yet iron ore supply is expanding massively. This investment is predicated on the assumption that Chinese steel/iron ore demand would necessarily grow rapidly along with its economy, but of course this is nonsense; if you are building a hundred new apartment blocks in your city each year, your city might growing rapidly but your annual demand for raw materials does not. Methinks iron ore miners, especially the weak hands, are in a shakeout.

        • As I’ve said the Steel mills I know have been working on $80 for the long term price for quite some time. I’m just passing on the anecdote not making any prediction.

          As you say MJV there is a big potential for overshoot here!

    • Troubling for the Oz economy if that proves the case. That’ll be the ToT shock from hell – what about coal? Majors OK (Vale…?) – plays well(ish) for them, eliminating range of producers. Opportunity for acquisitions, global growth will resurge at some point.

      Frankly can’t see IO settling at that price point med-long-term. Still see China muddling through, confident on infrastructure front and re commodities in the Grantham camp – long everything in the ground.

      Time will tell.



      • Thanks for the links 3d1k, but there is no way I’d believe this now I’ve seen the FMG Q report, and the research on the EV. The other miners might be ok, but I’m suspicious of FMG now.

        I got sucked on Aussie gold producers until I say the EV cost and the correction that gives you. I can’t remember all the numbers now, but I can think of a few that were 100% in error if you fall for the cash cost metric.

      • Do you know the projections for IO expansion in Africa? Obviously you can’t aggregate IO production costs for the entire continent, but if they really are sub-$40/t for the most part, couldn’t that additional supply conceivably mop up a lot of the supply lost as higher cost producers exit the market?

  5. @ 3d1k

    “Frankly can’t see IO settling at that price point med-long-term. Still see China muddling through, confident on infrastructure front and re commodities in the Grantham camp – long everything in the ground.”

    You are confusing the creation of the quantity money with the extraction of the quantity of IO!

  6. Just a single observation of no particular relevance.

    I was in China for a couple of weeks about a month or so ago. Everywhere I went we sat on the tarmac at take-off waiting for the line of planes, usually 5 or 6, in front to take off. On leaving the Aus pilot flying for Silk Air announced a delay and said that Airspace over China was now very heavily ‘occupied’ I forget the word used.

    Now China has been ridiculed for its Fast train build-out. Imagine what it would now be like without the Fast Trains they have built.

    We landed in Xiamen. Next day we were picked up to be taken to another ‘town’ of about 1.5 million people an hour or two away. We made the comment that when the Fast Train was built we would be able to catch the train rqather than put them to the trouble of picking us up in Xiamen. The co-ordinator from the factory said ‘Oh the Fast train will be no good for you’ We are very experienced Chinese travelers and can get by in the language to get from place to place. So we didn’t understand so we tried to reassure her we would be OK. She replied it might be OK after 6 months or so but that it would be too crowded and we would not be able to get a ticket at least for the first 6 months.

    I relate the story only in terms of there is a lot going on there that we have trouble comprehending. In that case it is pretty hard to make good judgements particularly from the point of view of the big flat screen in teh lounge in Aus.

    • dumb_non_economist


      China’s airspace is congested because of restrictions applied to its use for civil purposes. Aircraft are limited to narrow corridors with most airspace designated as restricted military use. Also prone to close these corridors with little to no notice for “security” reasons. Opening it up would make it more efficient and a better competitor to rail.

    • i don’t know, Chinese will do everything as long as it suits their own interests (financial).

      I landed to Beijing airport at approx. 7.00am and the same day at 12.15pm was sitting in the high speed train to Shanghai. (only 3 weeks after its official opening and 1 day before the deadly accident south of Shanghai). No online booking available then. Here is my journey: http://www.youtube.com/watch?v=PVKUvE2ny40

      5 days later when I was leaving Shanghai the hotel assistant receptionist and his friend taxi driver were showing me a number of newspapers with pictures of the train wreckage telling me trains were unsafe so they would get me to the airport by taxi avoiding dangerous Maglev (for only twice as much money)

      • Thanks alien…yes different stories from different perceptions!
        The place has fascinated me constantly over 28 years of going there.

  7. Crikey!!! Sorry about the missing punctuation…brain works slightly faster than fingers thank heavens!

  8. This was obviously written by a bloke a little concerned about keeping his job. From his perspective I think he has done a cracking job and guaranteed himself a cup of tea with Julia.

    • Going by the promotion handed to an individual who got the pre-GFC US financial system stability so badly wrong, I would say the authors will get their promotion soon.

  9. There is absolutely NO chance, none, that China will come close to 8% pa growth for next 20 years.

    Is there no limit to these peoples stupidity? Take a large company (China) and compare it to a smaller company (S Korea). Which has the best chance to grow faster longer? There is an inherent limit to the ability of large, complex entities to continue to grow. Not only that, the smaller entity was starting from a much better base. This is basic common sense yet it seems to have no place in RBA models.

    China has immense social, structural, demographic and geopolitical challenges which will choke off growth starting about now.

    A far better comparison would be the USSR in the late 80s/early 90s. Why is that never considered?

  10. H + H
    Where do you get 5.3% from?
    The High Growth Scenario is 8%
    Baseline Scenario 6%
    Low Growth 3.3% (2.7% below baseline)

    I think you subtracted 2.7% from 8%, rather than from the baseline of 6%

    I think from reading the report they foresee EITHER the baseline or High Growth Scenario occuring – they haven’t committed to either. Both of these scenarios see steel consumption growing for more than 10 years, with the baseline scenario slowly dropping away soon after 10 years.

    Personally I’d go for circa 4-5% GDP growth in the next 20 years ie. between low growth and baseline scenario.