Interpreting the news

I want to talk a bit more about the Chinese “bubble city” of Ordos, but do so indirectly, by discussing how I interpret news stories.  (First read the previous post.)  Unlike many other people, I don’t like to interpret news as “narratives,” instead I like to think in terms of numbers, and highly abstract economic principles.  This often makes me a contrarian. 

There have been some recent studies suggesting that right now the world is less violent than it has ever been in all of history, and even pre-history.  I have no idea whether the studies are correct, but I find them very appealing, as they seem to so dramatically contradict the “impressions” we get from watching the evening news.  And yet they do fit in with economic theory.  You’d expect violence to gradually become a less appealing option for people as the world got richer and life expectancy increased.

The video about the empty Chinese city outside of Ordos presents some stunning pictures that trigger all sorts of stories, or narratives, in our minds.  But this “manipulation” just makes me all the more determined to dig deeper, to try to find alternative interpretations. 

Even someone ignorant of economics would form all sorts of instant impressions from the sight of an empty city built to hold 1 million residents:

1.  Folly; the absurdity of building a city without residents

2.  Hubris; the belief that the Chinese government could get people to live in the middle of nowhere

For those who have studied economics there are even more narratives:

3.  Bubbles; another real estate bubble in the desert, even worse than Phoenix and Las Vegas

4.  Communism; the inefficiency that results from central planning

5.  Misallocation; a neo-Austrian vision of business cycles

and many others.

Although I am no China expert I have visited China 6 times, and traveled to many different regions.  So I also thought of other issues when viewing the video: 

1.   The people who live in the wealthier parts of China’s east coast tend to look down of the interior of the country as being very poor and backward.

2.  Even 15 years ago most Chinese construction looked very ugly and slipshod.

3.  Many Chinese people don’t like peace and quiet, they like to live in bustling cities.  (Obviously a crude stereotype, so let’s just say compared to Americans.)

The first two points led me to be pretty impressed by what I saw.  When I first visited China’s interior in 1994, something like the new city of Ordos would have been inconceivable.  The local government would have lacked the money and expertise to pull it off.   Just the fact of its existence points to the fact that Inner Mongolia must have made enormous economic strides in the past 15 years.  Those houses may look repetitive and bland to Americans, like the endless suburbs of the Southwest, but they look pretty impressive to the average Chinese citizen.  The third point listed above, however, makes me view it as even more of a boondoggle than the average American might think.  Lots of Americans like to buy houses in quiet suburbs 30 km from the hustle and bustle of downtown NYC or Chicago.  That is much less true in China.  (Although just as in Europe, you are beginning to see wealthy people moving to suburbs.)

Beyond these impressions, I like to think in terms of hard theory and hard numbers.  Here are three questions:

1.  Was the investment as bad as it looks?

2.  If it was as bad as it looks, how typical is it of current trends in China?

3.  If it is typical, how serious of a macroeconomic problem is it?

1.  I think it probably was as bad as it looks, although the fact that most units have been snapped up by investors makes even that seemingly obvious point a bit uncertain.  One person who was interviewed said that the prices were too high; people couldn’t afford to live there.  OK, but if the speculators are losing money, won’t they eventually cut the prices to the point where people can live there?  And if there is a sort of Chinese cultural agoraphobia, is it possible that a tipping point might be reached, where there is enough liveliness to the place so that it then rapidly fills up?  This is probably wishful thinking on my part, but I don’t think we know enough to rule out the possibility.  After all there were also building sprees in the 1990s in places like Shanghai that were reported by the American press as being “bubbles,” but turned out to be the exact opposite.  (As the guy interviewed in the video notes, few people have lost money in Chinese real estate.)

2.  I am much more dubious of the second point, the argument that Ordos is representative of China.  I won’t rehash everything from the previous post, but my impression is that it is atypical.

3.  Even if it is typical, does it mean what we are led to believe it means?  It is hard to see those pictures without thinking about the recent sub-prime fiasco in the US.  But Chinese banks don’t make sub-prime loans, so it is not at all clear that loan defaults will rise to US levels.  I also don’t know the relative importance of the government and the private sector in Chinese housing.  But I do know that some of China’s wealthiest people have gotten rich through property, so I assume that the private sector plays a significant role (as does the government.)

So we need to think in terms of what is the “worst case?”  Suppose Ordos really is as bad as it looks, and suppose there are lots of other places around China with significant overbuilding of housing.  What does that mean?  Well assume the US were to have overbuilt its housing stock by 10% (in fact our bubble was nowhere near that large.)  If so, it would have taken us years to work off that excess.  But in China the demand for housing is skyrocketing, and that excess would be worked off very quickly (except perhaps in a few extreme cases like Ordos.)

But let’s also suppose I am too optimistic about Chinese housing demand, and suppose there turn out to be massive loan losses for Chinese banks.  What then?  Well China experienced exactly that problem in the 1990s, and there were many rather apocalyptic statements made about Chinese banks.  I seem to recall that in the early 2000s China’s government did some banking reforms about the time it joined the WTO, pumped in some additional capital, and listed the banks on the stock exchange.  I presume another bailout is coming at some point.  Am I suggesting this is a good thing?  Of course not. Rather I am suggesting that as Adam Smith once said:

There is a great deal of ruin in a nation.

By moving away from communism China unleashed enormous growth.  Even if the Chinese economy is like a 8 cylinder engine where a couple cylinders aren’t firing, it can still grow very fast until it reaches a far higher per capita GDP.  (It is still poorer than Mexico.)   And that’s even assuming no further reforms.  But I believe that further reforms are very likely.

So these are the sorts of questions I ask when I try to put a story like Ordos into perspective.  I get the same impression as anyone else does by watching that video.  And it is very likely that the impression one draws is at least partly correct.  But the implications one draws about the Chinese housing sector as a whole are much less likely to be correct.  And even if those implications are correct, the implications that the Chinese housing sector have for their banking system, and the broader economy, may be incorrect.

One commenter asked me whether I was misusing Say’s Law.  Of course Say’s Law doesn’t settle any argument, it doesn’t prove any point.  But thinking in terms of Say’s Law can help us clarify our views.  For instance, Say’s Law may be violated in a cyclical context if the economy overheats and output exceeds the natural rate.  This essentially means you produce too little of the leisure good, and too much of all other goods.  But is this China’s problem today?  Obviously not, tens of millions of jobs were lost in global meltdown.  So then the argument must be sectoral, you have too much of some goods and not enough of others.  And that obviously is a problem in China.  But if you look at the Shilling and Dow videos back to back, it is also obvious that Dow has a more sophisticated understanding of the current boom in China, which is directed much more toward the domestic economy than the export economy.  Rather, the misallocation of goods in China is pretty much what you’d expect from a half-communist country—too much spending on big showy projects, not enough consumer services.  But even if Chinese growth was more balanced, there would still be rapid growth in spending on big projects, just somewhat less rapid than what is actually occurring.  As Dow indicated, 10% RGDP growth will quickly cover up a lot of resource allocation mistakes.

Part 2.  Chimerica

I believe that the historian Niall Ferguson coined the term ‘Chimerica’ to denote the emerging global duopoly of the US and China.  Many news discussions focus in the pathologies of this relationship, sometimes describing it like a dysfunctional married couple that is unable to live without each other despite the fact that each country’s policies create problems for the other.  I think these narratives are all wrong.  I don’t see the US/Chinese economic relationship as being particularly dysfunctional, or even important.  So why do so many believe otherwise?

1.  One mistake is the psychological tendency to overemphasize the importance of the biggest object in any group.  Back in the 1980s we often read reports about how “Germany” (then a country of only 62 million) was the key to the EU economy.  But this was silly.  Germany probably had about 25% of the EU GDP, but France Britain and Italy each had nearly 20%.  So there was really nothing very special about Germany, it just gave economic reporters a “story” to tell so that they looked more sophisticated.  Readers want the picture simplified; they don’t want mind-numbing numbers about the percentage of EU GDP produced in Holland, Belgium, etc.  And the German “locomotive” was a nice metaphor.  To the extent there was any truth to the story, it should have focused on the capital goods sector, which is indeed more cyclical than other sectors, and is somewhat over-weighted in Germany.  But that’s another (and less interesting) “story.”

The US/China relationship seems the same.  Yes, we do lots of trade with China.  We buy from China stuff we used to buy from Korea, Taiwan, Mexico, etc.  Big deal.  (Indeed lots of our “Chinese” purchases are actually stuff made elsewhere in Asia, and simply assembled in China.)  We also still do lots of trade with Canada, Mexico, and many other countries.  Yes, China buys lots of Treasury bonds, but so do many other countries.  And why is it important that the Chinese buy lots of Treasury bonds?  The implication is that something bad would happen (for them, or us, or both) if they stopped buying lots of Treasury bonds.  But what is that bad thing that would happen?

1.  One possibility is that China keeps running big trade surpluses and uses the funds to buy other assets.  Perhaps they could buy various Japanese and European government bonds.  OK, then the price of Treasury bonds falls a bit and the price of alternative bonds rises a bit, and some bonds are swapped between investors.  Who cares?  It’s not like the prices would change all that much, after all, various government bonds are close substitutes.

2.  Another possibility is that China stops running a big trade surplus.  So now we need to think about whom would then buy these bonds (assuming other countries don’t pick up the slack.)  I suppose the fear is that the Treasury bond prices would have to fall, and interest rates would have to rise, until Americans could be induced to increase their savings enough to buy these bonds.  And (so the argument goes) the extra saving and higher interest rates would depress the economy.  But it isn’t clear whether this is supposed to be a cyclical or secular problem.  Here is where rigorous theoretical analysis is crucial.  What exactly is the problem we would allegedly face?  Does the increased saving put us into a recession?  Remember that we are assuming China and the US are moving closer to balanced trade.  So doesn’t the negative of more saving get offset by the positive of more net exports from the US?

A more sophisticated argument might use some sort of Keynesian model where the net effect is lower aggregate demand (or lower velocity.)  I’m not quite sure how higher interest rates on Treasury bonds would lower velocity, but let’s suppose there is some mechanism.  In that case all we have is a cyclical problem, not a secular problem.  And if the monetary policymakers are at all competent, we don’t even have a cyclical problem, as they could print enough money to prevent the higher interest rates from slowing AD. 

As we saw between mid-2006 and mid-2008, when resources were reallocated out of housing construction and into export industries, there isn’t much impact on the US unemployment rate when the tradable goods sector of the US expands.  Only when NGDP falls do we have a major cyclical problem.

To conclude, these Chimerica stories sound very appealing.  They appeal to the deep human instinct to arrange boring and abstract facts into appealing narratives.  But a closer look at the numbers involved and the relevant economic theory shows that these stories are completely vacuous, just a bunch of fairy tales about current account “day of reckonings” that never seem to arrive.

And the Chinese development stories are also appealing.  They appeal to our wanting to know “what’s really going on in China.”  There is no single narrative “really going on in China.”  There are 1.3 billion narratives going on, that can only be understood through careful analysis of the relevant numbers, and the relevant abstract economic principles.

Soon I’ll explain why despite our negative impressions of India, formed through media stories, it will have the world’s biggest economy in 100 years.



19 Responses to “Interpreting the news”

  1. Gravatar of Thorfinn Thorfinn
    18. November 2009 at 08:38

    It seems to me that the “G-2” is seen as important exactly because it is seen as a problem (even if it isn’t)–on currency, climate change, etc. But it is unlikely to work as a unit of global governance exactly for that reason.

    Looks like down payments for real estate are around 40%. So as others have said, the bigger problems are with loans to SOEs. Maybe China will deal with this problem effectively; but the experience of other East Asian countries with similar asset bubbles doesn’t inspire confidence.

  2. Gravatar of jimz jimz
    18. November 2009 at 09:42

    See Tyler Cowen on narratives at Ted:

  3. Gravatar of Doc Merlin Doc Merlin
    18. November 2009 at 10:27

    I don’t think Say’s Law applies here in to the Chinese case.

    Demand and supply are the same thing. Obviously, one person’s demand is their trading partner’s supply. (e.g. so one person’s demand for TV’s is their supply of money and vice versa for who they are a buying a TV From.)

    This would imply that any expansion of the money supply increases the demand for money over the long term, this makes me uncomfortable with the formulation of Say’s law that you use. It seems to me that Say’s law would only apply to situations where demand side market expectations of future supply get fed back into the supply side, like in production of goods, or, as I have discussed here before, money with futures’ convertibility. But it wouldn’t apply to things like socialist central planning or to fiat money.

    Chinese government’s fiscal stimulus for building cities has this same problem. Sure it increases housing/city supply and it shows great architectural feats, but it wasn’t built based off of market expectations. It was built by government central planning which is insensitive to market expectations.

  4. Gravatar of rrm364 rrm364
    18. November 2009 at 10:35

    first, in response to the post :

    I think a big part of why coastal Chinese think that the interior is backwards is simply because the interior simply is dramatically poorer. Per capita income in roughly eight times higher in Shanghai than in Guizhou. Even when comparing non-states to Chinese states without substantial minority populations you see the same thing. Per capita income in Zhejiang is $11,428 PPP while it is $3,278 PPP in Gansu. Moreover, non-performing loans china npls and many Chinese banks would be insolvent without the promise of a government bailout.

    In general, I’m a lot more pessimistic about China than you are. Many countries, like Brazil, Mexico and Cote D’ivoire have developed rapidly from lower to upper middle status, but only a handful have gone from upper middle to upper income. This isn’t exactly high quality econometrics but I looked at countries with a PPP per capita income between 1/2 and one 1/4 of the US in 1980. Almost all of these countries (except for Taiwan) fell grew slowly than the US during the next 30 years and in some cases the differences were dramatic.
    The speicific constraints that I think China faces is a bank system that cannot channel resources to where they are needed most (which is a big reason why corporate savings in China are spectacularly high). SOE have a huge unfair advantage over their private domestic competitors and this disadvantage may even be increasing because of the nature of the Chinese stimulus. At some point, the Chinese government will have to bailout its banks and these banks could place huge fiscal costs on the chinese economy. More worryingly, Chinese companies will need to be able to compete on both cost and quality, something SOEs have generally struggled to do without subsidies.
    Another problem is the inability to produce human capital. My guess is the explosion in inequality has a lot to do with the fact that the demand for education (both quantity and quality) are going up much faster than the supply. I suspect like Brazil and Mexico, China will soon (by soon I mean within the next 20 years) go through an extended period of stagnation.

  5. Gravatar of rrm364 rrm364
    18. November 2009 at 10:45

    Oh, one unrelated thing. I’m an undergrad student doing some research on forex reserves and undervaluation. I saw some people people use forex reservse/ GDP to see whether in some countries (esp China) forex accumulation is a sign of deliberate undervaluation. However, I don’t see any theoretical validity to putting gdp in the denominator. My adviser suggested I use base money, but wouldn’t M2 be a better measure. Plus, I’m still not clear on the theoretical validity of either measure. Sorry about the unrelated request, but I’m wondering if anyone has any suggestions.

  6. Gravatar of David C David C
    18. November 2009 at 10:58

    I’d be very interested in your take on India. The other two major developing countries (India and Brazil) simply don’t get enough coverage. On the surface, it seems unlikely that an economy that is currently less developed and has fewer people than China will suddenly outpace it.

  7. Gravatar of 123 123
    18. November 2009 at 12:57

    What would Hayek say?

    1. China is on a road from serfdom
    2. Road from serfdom has many detours, one of them is Ordos
    3. When China returns to the main road, the question is which direction will it turn to?
    4. If enough people read Hayek in China, the direction of China will be right

  8. Gravatar of Bill Stepp Bill Stepp
    18. November 2009 at 14:02

    David Dreman: Double digit inflation and interest rates coming
    (Don’t tell PK.)

    A hot shot CEO of a startup international courier firm told me recently that pads of note paper peddled by Staples, etc. will not exist in ten years. Yeah, right, we’ve been hearing about the coming paperless office for the better part of 30 years.
    I like Dreman’s bet a lot more.

  9. Gravatar of Scott Sumner Scott Sumner
    18. November 2009 at 15:46

    Thorfinn, Thanks for the housing downpayment numbers. I thought it relevant as the video said most of the units had been sold to speculators. So that is their problem, not the Chinese government’s problem. SOE loans are a problem, as you say.

    Thanks jimz.

    Doc Merlin, I’m not quite sure how you object to my use of Say’s Law. I never brought money into the picture.

    rrm364, I don’t think it makes sense to compare China to Latin America and Africa, the cultures are too different. A better comparison is other East Asian economies, especially those with Chinese ethnic majorities (HK, Taiwan, Singapore.)

    I agree that China will bail out the banks. They’ve done it before. I don’t see that as a big problem for a high saver like China. I expect them to continue reforming their economy, so the banking system will gradually become more efficient.

    I can’t help you with the reserves question, as I don’t think the foreign reserves totals are very useful indicators of currency undervaluation, no matter what denominator you use. I’m not even sure what the term “undervaluation” means. Presumably there must be a benchmark as to the proper valuation, so if you have that benchmark why bother with reserves at all?

    David C, Not suddenly, in 100 years. See my new post.

    123. Your comment (which I agree with) reminds me that I should have mentioned than in China something like Ordos is simply not a big deal. Most Chinese have probably never heard of it, and most probably don’t care. Inner Mongolia might as well be the moon as far as many Chinese are concerned. And a million people sounds like a lot, unless you have 1.35 billion.

    Bill, I think 1% inflation and 2% interest rates are coming.

  10. Gravatar of Scott Sumner Scott Sumner
    18. November 2009 at 17:07

    jimz, I just watched the video. It was so good it makes me embarrassed to even be addressing the same issue. I should have just linked to Tyler.

  11. Gravatar of Frank Frank
    18. November 2009 at 17:14

    Can someone educate me about the monetary policy impact of China hording US dollars, i.e. they are pegging their RMB to the US Dollar by buying up dollars and now hold about 2.4 trillion US dollars in reserves. I see this as offsetting the US Federal Reserves’ so called QE policy (which I consider to be weakly defined in the business press). But in layman’s terms (I am a layman), it seems we have tight money on main street and that the Fed may be “printing” money but that China is “hording it” and offsetting the FED’s efforts. Help

  12. Gravatar of OGT OGT
    19. November 2009 at 06:24

    Pettis makes the same point in regard to the infrastructure. He also notes that there is a great deal of duplication, investing trains and brand new regional airports meant to seve the the same routes.

    The recent history in China is for investment losses to be socialized out of direct or indirect subsidies from the household sector to banks and firms. This is troubling given that the level of consumption as a percentage of GDP is already at record lows (estimated at 35%).

    One, I think, also has to take into account how truly huge the amount of investment being made in China is right now, perhaps a modern record as a percentage of GDP.

    On the one hand, with such a huge amount of investment it would be unfathomable that some large scale waste didn’t take place, so specific examples should be put in perspective. On the other, the share of GDP means that the bet(s) are a big gamble that China can ill afford to lose.

    Ordos is somewhat disturbing, but to me, less so than the reports of leveraged iron and copper hoarding by SOE’s being doled out ‘loans’ they have no productive outlet to invest.

    I am eager to read your take on India. India doesn’t seem to get enough coverage of any kind.

  13. Gravatar of rob rob
    19. November 2009 at 10:43

    You are too Hegelian. The immense progress humanity has made over the past millennium was nothing but a super bubble. World GDP in 100 years will be less than today, as it reverts to the mean. (Please see my book: “Shorting Stocks for the Long Run”).

  14. Gravatar of rob rob
    19. November 2009 at 11:42

    that was meant to be in response to your India post.

  15. Gravatar of Anonymous Coward Anonymous Coward
    19. November 2009 at 22:55

    I was saying that Say’s law would only apply to things that have some way of feeding back market future expectations into influencing the shape of their supply curve.
    The normal way producers get expectations input from the rest of the market is through the supply (and the costs) of the things a producer needs in order to produce his goods.
    Where a producer’s expected marginal costs are very, very low compared to their expected marginal income or their fixed costs, you can get oversupply. A good example of this kind of oversupply was online service CD’s and free trial subscriptions in the 90’s. I knew several people who who had no actual subscription but would be online all the time because of so many free trail offers.
    This also happens when a producer is insensitive to costs like is the case with the US or Chinese governments. They are insensitive to spending costs, because they can get income at very little marginal cost. This means that fiscal stimulus often causes oversupply.
    My example with the Federal Reserve causing oversupply was to link an example of why I think that the lack of market expectation information doesn’t just apply to fiat currencies. It also applies to This is why I don’t think Say’s law applies to things like Ordos. Ordos may very well eventually fill up, but I don’t think it necessarily has to.

  16. Gravatar of Doc Merlin Doc Merlin
    19. November 2009 at 22:56

    Heh, well if you can’t tell I’m Anonymous Coward, I thought the name was funny (from so I’d try it on for that one comment. Forgot to clear it though.


  17. Gravatar of David Heigham David Heigham
    24. November 2009 at 08:49

    Speculative house building alway produces, from time to time, developments which sit laregly empty for a while. (The most spectaular of the current crop here in Spain is outside Sessana, not far from Madrid.) The years before they fill up, and the eventual sales prices, pose problems for those who have financed them; but over a longer perspective they do not constitute major resource waste.

    More seriously, about China, there are a billion small narratives now going on in China – that is good news however we look at it. But there is still one big narrative going on in China – the narrative of the Communist Party/mandarinate. That big narrative has so far been astoundingly (as compared to ther Communist regimes and to past Chinese history) benign both to its citizens (I surprise myself, it now makes some sort of sense to talk of individual Chinese as citizens rather than as just passive subjects of the political regime.) and to the rest of the world. But the big narrative still has great power to mess up the small narratives and the rest of the world. I would be much more relaxed if I thought that the big narrative could “be understood through careful analysis of the relevant numbers, and the relevant abstract economic principles.”

  18. Gravatar of rob rob
    24. November 2009 at 20:51

    good economists dont make predictions, unless they are very long term. extrapolating the past rarely works. why not 100 years ago extrapolate that china or india would have the biggest economy today? why not austria-hungary?

    i do think the whole theory is too panglossian. the next 100 years is likely too see more world wars and major revolutions.

    i dont know how to make smiley faces. i am always posting on my phone from some airport bar.

  19. Gravatar of Scott Sumner Scott Sumner
    25. November 2009 at 17:09

    Frank, No, the Chinese aren’t holding any US dollars, they hold Treasury bonds and other financial assets, so it doesn’t impact US monetary policy.

    OGT, I mostly agree, although I would point out that the main problem isn’t so much the building of airports and rail they don’t need, but rather building them too soon. I also think that consumption is much higher than the official figures show, although the bias toward investment is real.

    rob, I am beginning to detect a bubble in bubble theories.

    Doc merlin, I don’t think oversupply is the problem in China, they could use much more of almost everything (cars, homes, home appliances, services, infrastructure, etc.) The main problem is inefficiency of production. Privatization would help a lot.

    David, Good point. In an earlier post I argued that it was very good news that most Americans didn’t know the name of China’s leader. In the 1960s (when China’s government was much less benign) most American’s did know who Mao was.) Let’s hope Americans continue to be ignorant of the name of China’s leader, it is a sign of gradually increasing political maturity in China. But I agree that there are still substantial risks–they have a long way to go.

    rob, I don’t expect any more world wars. I don’t expect many more big revolutions, although there may be a few more here and there. I still think history is ending, and the new threats will be from terrorists and/or scientists.

    For a smiley face, just hit colon then right parentheses:


    But I was kidding about the smiley faces, just making fun of myself. Don’t feel obligated.

Leave a Reply