Either Lars or I am really, really wrong

I’ve argued that China is currently overtaking the US in the race to have the world’s largest economy, and will have an economy twice the size of the US within a few decades.  I’ve picked on Lester Thurow for claiming it wouldn’t happen until the 22nd century.  And now Lars Christensen has suggested it might never happen.  Those are fighting words to a Sinophile like me.  Here’s Lars:

However, the [World Bank] argument was completely bogus as it was based on Purchasing Power Parity (PPP) rather than on actual exchange rates (To be fair we should blame the media rather than the World Bank for this interpretation of the data).

PPP based measures of GDP (per capita) might make sense if we want to measure how much an average citizen can buy for given an average income, however, it does not make sense when we want to measure the size of the economy. There we have to use measures based on actual exchange rates and if we do that then it turns out that the Chinese economy is still significantly smaller than the US economy. Hence, total Chinese GDP today is around 10 trillion USD, while US GDP is around bn 17-18 trillion USD. Said in another way the US economy is still nearly double the size of the Chinese economy.

And what I will argue in this post is that China might never overtake the US as the biggest economy in the world.

I have several problems with this argument.  First, if Lars really feels PPP is wrong, and that we should use nominal figures, then he should not be talking about China having recently grown at 7 to 7.5% per year.  In PPP terms China may have been growing at 7.5% vs. 2% in the US, but in nominal terms the gap is far wider, due to the Balassa-Samuelson effect.  China’s real exchange rate has appreciated strongly over the past decade.  So if Lars is right that nominal exchange rates are the right test, then China’s been catching up to the US at a rate far faster than either Lars or I assume. And in that case China’s nominal growth could slow dramatically and yet still be growing far faster than the US (where trend NGDP growth is now about 3%, in my view.)  Lars avoids this problem by assuming the Balassa-Samuelson effect will suddenly come to a screeching halt, whereas I think the yuan is headed to 4 to the dollar.  He also assumes a 3% RGDP growth rate for the US, whereas I believe it will be closer to 1.2%, growing over time to perhaps 2% in a few decades.

My second argument is that it makes no sense to use nominal exchange rates to compare the size of economies.  Perhaps it makes sense if you want to look at the impact on global trade (China exports far more than the US, BTW), but surely not if comparing domestic production.  I recall the euro being about 85 cents around 2002, then by 2008 it peaked around $1.60.  And yet the US and eurozone had roughly similar NGDP growth rates over this 6 year period.  Does anyone believe that comparing non-PPP adjusted GDPs would have given a meaningful comparison of the relative size of these two economies?  Did the eurozone suddenly go from having an economy much smaller than the US to one far larger, in six years?

And consider the effect of tax regimes.  Suppose you adopt a VAT that provides revenue equal to 20% of GDP.  Your nominal GDP at market exchange rates will suddenly jump by 20%, even though nothing has happened to the real size of your economy.  Indeed the European VATs are one factor that explains why Europe often looks better against the US if you don’t adjust for PPP.

Switzerland’s nominal GDP per capita is $81,323, and Germany is $44,999. Here are three more:

Greece   $21,857

Taiwan, $20,930

Portugal,  $20,727

Now let’s go to PPP.  We find Switzerland at $46,430, and Germany at $40,007. Doesn’t that gap seem more plausible?  In PPP terms Taiwan has a bit under $40,000, whereas Greece is just over $24,000 and Portugal a bit over $23,000. Again, far more plausible.

Still not convinced?  China produced more cement in 2011-12 than the US did in the entire 20th century.

Still not convinced?  Think about the fact that China has 1.36 billion people while you read this:

CHINA Northern Locomotive and Rolling Stock Corporation (CNR) has been awarded a contract to supply a fleet of 60 driverless trains for the first metro line in Beijing to be equipped for Unattended Train Operation (UTO).

I’m telling you, this country isn’t messing around.

For Lars to be right, China’s nominal GDP per person would have to get stuck at Brazilian levels.

Ladies and gentlemen of the jury, I rest my case.

PS.  Under PPP Denmark drops from over $59,000 to under $38,000.  Maybe that’s why Lars likes the unadjusted figures.  🙂



64 Responses to “Either Lars or I am really, really wrong”

  1. Gravatar of David R. Henderson David R. Henderson
    28. August 2014 at 06:15

    You’ve convinced me: Lars “am” wrong. 🙂
    One thing I don’t get, though, is why you say imposing a VAT that takes 20% of GDP would increase GDP by 20%. Doesn’t that assume that the supply curve of output is vertical.

  2. Gravatar of benjamin cole benjamin cole
    28. August 2014 at 06:18

    Seems likely the Han Chinese will migrate to Japan-level incomes in 20 years. The question is, will a very long term stagnation set in somewhere in China’s future, ala Japan or now Europe.

    The PBoC is growth-oriented, a huge plus. But politically, the society is becoming more backward and repressive. And remember, CCP controls every public company through voting stocks or board seats. Ergo, a longish stagnation period may be in the cards.

  3. Gravatar of ssumner ssumner
    28. August 2014 at 06:21

    David, I suppose I am assuming the long run effect, where the LRAS is vertical. I agree that short run effects may differ.

    Ben, I think they’ll get there, but it will take more than 20 years.

    I strongly disagree with the claim it is becoming more “backward.” China is much more liberal than when I visited in 1994.

  4. Gravatar of Trevor Trevor
    28. August 2014 at 06:39

    … Vancouver has had a driverless metro system since 1986. Strong transit unions prevent a wider adoption in other cities, not technical capacity.

  5. Gravatar of Luis Pedro Coelho Luis Pedro Coelho
    28. August 2014 at 07:39

    Lisbon has had driver-less subway trains for over a decade now, but a driver must be sitting in them at all times (for emergencies, you know). When drivers go on strike, the driver-less trains do not run.

    Budget cuts have meant that the driverless trains run less frequently, but they still have a driver.

  6. Gravatar of Steven Kopits Steven Kopits
    28. August 2014 at 07:40

    IMF has China overtaking the US in 2019, if I recall correctly.

  7. Gravatar of Andrew M Andrew M
    28. August 2014 at 07:41

    China’s working age population began shrinking in 2010; in 2020 the total population will begin shrinking. The dependency ratio will soon become ghastly; worse than Japan levels.

    China will become the first country to get old before it gets rich. How exactly is it supposed to surpass the US with such demographic headwinds?

  8. Gravatar of collin collin
    28. August 2014 at 10:28

    Although I agree that the Chinese economy will overtake the US at some point, but isn’t Chinese economy just taking a bullet train to the Japanese crash and the last two decades of economy.

    Build a huge economy around exports and overwork the population to pointthat nobody has large families. At this point, doesn’t it appear without enormous government interventions, the Japanese economy is a giant demographic deflation spiral?

  9. Gravatar of Jason Smith Jason Smith
    28. August 2014 at 10:36

    A prediction for China: above 10% NGDP growth (on average) to 2020. Slowly falling RGDP growth, averaging about 8% by 2020. Some inflation predictions and other graphs here:


    China appears to be entering a period similar to the US in the 1960s, but without having had a Great Depression or WWII earlier.

  10. Gravatar of jj jj
    28. August 2014 at 10:46

    Scott, can you explain the VAT-increasing-NGDP thing a bit? I usually follow what you’re saying but that went right over my head.

    I would expect that to maintain output after a 20% price increase due to VAT, the central bank would have to increase money supply by 20%, causing the exchange rate to rise by 20%. Therefore the NGDP at current exchange rates would be unchanged.

  11. Gravatar of mikef mikef
    28. August 2014 at 10:54

    “CHINA Northern Locomotive and Rolling Stock Corporation (CNR) has been awarded a contract to supply a fleet of 60 driverless trains for the first metro line in Beijing to be equipped for Unattended Train Operation (UTO).”

    There are some things that are so much easier to accomplish when people don’t have property rights or labor rights. But at some point…these become problems. Mass transit is a pretty clear priority in a country like China…but at some point the command economy is going to make big mistakes and cost economic growth. Yesterday’s story of the huge stockpiles of grains in China come to mind. These will be sold for huge losses. Add on to that the aging population, lack of resources, environmental degradation, rise of robotics/insourcing of manufacturing.

    If China were to adopt HKs government, then I would agree with you…and that may happen someday, but it is far from certain.

  12. Gravatar of ssumner ssumner
    28. August 2014 at 11:10

    David, I forget to mention, I is not very good at grammar.

    Steven and Andrew, China has already overtaken us; in 20 years China will have 4 times our population and twice our GDP.

    Collin, Japan has some of the highest living standards on Earth. A few decades back China was as poor as the Congo, poorer than India. If ending up like Japan is failure, I’d love to see success.

    JJ, No, a VAT affects the real exchange rate, as goods are now priced 20% higher than in the US (which has no VAT.)

    Mike, They’ve already caught up to the US in GDP, so no need to worry about whether that will happen.

    You said:

    “but at some point the command economy is going to make big mistakes and cost economic growth.”

    China has already had it’s command economy disaster, it was under Mao. Now they are trying to dig out from under the rubble, by adopting markets. I’d recommend Coase’s book on China.

  13. Gravatar of mpowell mpowell
    28. August 2014 at 11:11

    I think you are right about whether we should use PPP for this kind of measure. On the other hand, if you think you can use current growth rates and change in growth rates (whatever you think they are) to project relative size 50 years from now you are completely insane. I don’t think you are trying to do this, but it sounds like Lars is. The only question is whether the fundamentals underlying the Chinese economy (govt, demographics, culture, natural resources) compared to the US, will allow GDP/capita to close enough or not. Growth rates over a few years have way to much noise from other sources to think you can extrapolate over multiple decades just on that basis.

  14. Gravatar of brendan brendan
    28. August 2014 at 12:13

    People really need to stop equating demographics merely w/ how many ya got and how old they are. China’s aging, yes. But if projections hold, the ethnic composition of the US will change dramatically in the time-frame Lars examines. Far as I can tell economists simply assume that won’t matter, as if the US’s long-term growth rate in some intrinsic property of its coordinates on earth, independent of who lives here. That seems iffy enough that I’d like to see it defended rather than assumed.


    “I have indexed Chinese real GDP at 55 in 2014 – reflecting that Chinese GDP (in USD) is around 55% of US GDP. In my simulation I have assumed that US trend real GDP growth is 3%. This is probably slightly optimistic compared to the “consensus” among long-term forecasters, but it is basically the growth rate we rather consistently have seen in the US economy since the early 1960s. The American demographic challenges are somewhat smaller than is the case for China and I find it rather likely that the US gradually will adjust immigration policies so meet these challenges (I certainly hope so…)”

  15. Gravatar of JimP JimP
    28. August 2014 at 12:14


    Buiter on why helicopter money must always work.

    “The Simple Analytics of Helicopter Money: Why It Works “” Always”

  16. Gravatar of Dan S Dan S
    28. August 2014 at 13:20

    Actually I think “am” was correct. If you use “or,” then whichever thing comes after the last “or” gets to drive, so to speak. So it would be, “Lars or I am wrong” or it would be “I or Lars is wrong.” But I agree it still sounds awkward which is why people avoid that construction.

  17. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    28. August 2014 at 13:38

    Almost sounds like Brad DeLong is regretting voting for Obama;


    ‘A year and a half ago, when some of us were expecting a return to whatever the path of potential output was by 2017, our guess was that the Great Recession would wind up costing the North Atlantic in lost production about 80% of one year’s output-call it $13 trillion. Today a five-year return to whatever the new normal might be looks optimistic-and even that scenario carries us to $20 trillion. And a pessimistic scenario of five years that have been like 2012-2014 plus then five years of recovery would get us to a total lost-wealth cost of $35 trillion.

    ‘At some point we will have to stop calling this thing “The Great Recession” and start calling it “The Greater Depression”. When?’


  18. Gravatar of BC BC
    28. August 2014 at 13:47

    On the issue of PPP vs. market exchange rates, I am inclined to believe that market exchange rates are the “correct” ones. When calculating domestic GDP, no one would say that 12 apples should count the same as 12 oranges. Apples and oranges are different, and we combine them using market prices, the prices at which apples can be exchanged for oranges: $100 worth of apples count the same as $100 worth of oranges. Similarly, cement delivered in China is different from cement in the US. Market FX exchange rates determine how consumption in China can be exchanged for consumption in the US. If market exchange rates are such that 1 lb of cement in the US has the same USD price as 1.5 lbs of cement in China, then *both* Americans and Chinese would prefer 1 lb of cement in the US to 1 lb of cement in China: a Chinese person could sell the 1 lb of cement in the US, exchange the USD for yuan, and purchase 1.5 lbs of cement in China. Worldwide, everyone would agree that 1 lb of US cement is “worth” the same as 1.5 lbs of China cement, not 1 lb of China cement, as PPP would imply. So, China’s GDP is smaller than PPP would indicate because it is difficult to produce US cement in China.

    I would amend Lars’s statement to, “PPP based measures of GDP (per capita) might make sense if we want to measure how much an average citizen can buy, *ignoring consumer revealed preferences embedded in market exchange rates and prices*.

    Having said all that, when comparing China and US growth rates, one does need to be consistent in the prices used, e.g., compare nominal growth rates measured in USD at market exchange rates. I take no position on whether or when China will pass the US.

  19. Gravatar of Major.Freedom Major.Freedom
    28. August 2014 at 14:33

    Given the assumption of capitalism existing in Chinese territory and US territory, and given the assumption that China will have a larger population, then it is an absolute certainty that China will eventually overtake the US as the largest economy.

    The same reason why the size of the US economy minus the size of the Houston economy is larger than the size of the Houston economy, is the same reason why China will eventually have a larger economy than the US.

  20. Gravatar of W. Peden W. Peden
    28. August 2014 at 15:07


    It’s interesting to be reading about the contemporary views of certain economists in the 1980s (Minford, Buiter etc.) and to find out that they’re still publishing intriguing articles. Of course, that’s because (for those of us born in the late 1980s) the 80s are a lifetime ago.

  21. Gravatar of W. Peden W. Peden
    28. August 2014 at 15:11

    Incidentally, just to confirm everyone’s priors, back in 1983 Jeffrey Sachs thought that the early 1980s productivity revolution in the UK was a simple consequence of a high unemployment rate. (Because productivity is countercyclical?!) Sachs didn’t get that one right: the early 1980s was the beginning of a secular improvement in UK productivity in which we more or less caught up with the Germans (our ambition in so many things!) by the early Noughties.

  22. Gravatar of JimP JimP
    28. August 2014 at 15:43

    W. Peden –

    To me the 80s are an eyeblink ago. Or less.

  23. Gravatar of Garrett Garrett
    28. August 2014 at 15:59

    Either Lars [is wrong,] or I am wrong.

    The part in brackets is implied. Boom grammar.

  24. Gravatar of Mike Sproul Mike Sproul
    28. August 2014 at 17:31

    Ditto what Garrett said.

    Luck for you, me and Garrett is real good at grammar.

  25. Gravatar of ssumner ssumner
    28. August 2014 at 17:55

    This is getting embarrassing. So I had it right, and changed it to the wrong construction?

    Free market economists should not be language purists, should they? I mean languages naturally evolve, just like economies evolve. So why impose a rigid structure? It always seemed to me that “You be doing that” or “He be coming home” is more sensible, easier to learn than our current verb system. Yet it’s often associated with uneducated minorities.

  26. Gravatar of ssumner ssumner
    28. August 2014 at 17:57

    BC, You are missing the point. Prices differ because markets are segmented by barriers. In the case of China you don’t even have a free market in the currency.

  27. Gravatar of Garrett Garrett
    28. August 2014 at 18:06

    We’re all just signaling, professor. Ask Caplan.

  28. Gravatar of Rajat Rajat
    28. August 2014 at 22:53

    Hmm, I think I agree with Lars on your second point. Of course if we’re interested in living standards, PPP comparisons make sense. But if we just want to measure the raw value of the output of an economy, then why not use market exchange rates? After all, value is what markets tell us it is, isn’t it? I have no problem with the absolute and relative size of the European (or Australian ) economy rising or falling dramatically as exchange rates change. That’s the market telling us that the value of a widget produced in Europe or Australia is changing, just like the value of gold and other commodities can vary dramatically from year to year or minute to minute.

  29. Gravatar of mbka mbka
    29. August 2014 at 01:14

    I can anecdotally confirm that Denmark has a vastly overvalued currency right now. This summer, I found Munich, Stockholm, and Reykjavik, all cheaper than Copenhagen. Some by a wide margin. Not to mention Berlin or Vienna which are complete bargains re: Copenhagen.

  30. Gravatar of mikef mikef
    29. August 2014 at 04:52


    I think my quick reading of the post led me to think you are saying that the GDP/person of the Chinese will overtake the US…I realize the GDP of China is larger or will soon be larger than the US. That region of the world has so many people that it already dominates the world economy. But to really lead in terms of innovation, the government will need to further change. To me size is not as important as who is leading the human race to a better future.

  31. Gravatar of benjamin cole benjamin cole
    29. August 2014 at 06:10

    If we look at per capita GDP, we find China now around $10–12k and Japan at perhaps $30–35k. So if China grows by 7 percent annually per capita GDP, we will find China passing Japan in about 20 years (assuming Japan continues to stagnate). Also, I said Han Chinese, which eliminates some poorer inland zones.
    Maybe, maybe not.

  32. Gravatar of Dan S Dan S
    29. August 2014 at 07:35

    Linguists would agree with you. They say that there is no correct or incorrect usage of language. However people use it is just how it is and ought to be studied. English professors at Oxford on the other hand…

  33. Gravatar of Vivian Darkbloom Vivian Darkbloom
    29. August 2014 at 08:39

    Via Tyler Cowen: Are wages becoming less sticky because they are less like “wages”?


  34. Gravatar of TravisV TravisV
    29. August 2014 at 11:18

    Tyler Cowen:

    “Further evidence that the housing crisis was about screwy beliefs, not moral hazard”


  35. Gravatar of Lars Christensen Lars Christensen
    29. August 2014 at 11:50


    Do we agree that China’s GDP/cap relative to the US is where South Korea or Taiwan were in 1990? And then tell me what have happened the to real effective exchange rate in these two countries? Or Japan since 1970?

  36. Gravatar of Kevin Erdmann Kevin Erdmann
    29. August 2014 at 13:28

    Travis, don’t you love how Tyler and all of his commenters are just certain that the bust was inevitable, and that the sign of wisdom would have been to know in 2006 that the bottom was going to fall out? They just can’t even imagine that insiders were bullish because that was a reasonable position.

  37. Gravatar of TravisV TravisV
    29. August 2014 at 13:54

    Kevin Erdmann,

    This is a very complex topic, but I appreciate your view.

    I sense that the FDIC / Fannie / Freddie moral hazard did play some significant role in driving condo values too high in South Florida and home prices too high in Las Vegas / Riverside, CA. But tight money was far far far far far more harmful than that moral hazard.

    As Sumner has said, there was a cold, but tight money morphed the problem into serious pneumonia. http://vimeo.com/38915078

  38. Gravatar of AbsoluteZero AbsoluteZero
    29. August 2014 at 14:23

    Data from:
    IMF is used for all.

    1990: Taiwan/US = 33.81%, S.Korea/US = 26.38%
    2013: China/US = 12.71%

    1990: Taiwan/US = 41.38%, S.Korea/US = 32.86%
    2013: China/US = 18.54%

    12.71 is close to 26.38 or 33.81?
    18.54 is close to 32.86 or 41.38?

  39. Gravatar of Neal Neal
    29. August 2014 at 21:28

    To believe that China will *never* overtake the US economy is to believe that the Law of One Price will never manifest between the US and China.

  40. Gravatar of James in London James in London
    29. August 2014 at 21:43

    Good points. Both right.

    It was a good debate between insiders in 2006 and 2007 about housing peaking. The most bullish insiders reasonably argued HPI could never turn negative, because it never had. The bear’ish ones feared the leverage that had built upon that assumption.

    Neither side could predict that central bankers, now more officially tasked with complex banking regulation as a responsibility, might want to let markets “naturally” discipline that high leverage position. And most smart people agreed with the banking regulators.

    And at the same time it was hard to predict just how dogmatic central bankers, and smart people generally (including dumb people like me), had become about targeting (even) headline inflation. With hindsight only could you see what a dangerous cocktail was brewing up a horrible accident.

  41. Gravatar of James in London James in London
    29. August 2014 at 21:57

    It is possible Switzerland GDP per capita is nearer 2x Germany than that it is the same. Within Germany Hamburg is 2x the lowest states. Luxembourg is 50% more than Germany. Zurich and Geneva/Lausanne make up most of Switzerland and they are very rich.
    Cities and city states, HK & Singapore etc, as you often argue, are much wealthier than average. Switzerland on PPP vs Germany looks wrong.

  42. Gravatar of W. Peden W. Peden
    29. August 2014 at 22:32

    I don’t know much about the relative merits of nominal vs. PPP measurement, but I agree with James. Many parts of Germany are very backward (and not just in the East) while Switzerland is the biggest “super-prosperous microstate” in Europe. That whole area (Switzerland, Liechstenstein, Apline France, western Austria) has just exploded with development in every respect, and is full of high-wealth individuals. Even northern Italy is very dynamic.

  43. Gravatar of Kevin Erdmann Kevin Erdmann
    29. August 2014 at 23:33

    James in London,

    What leverage in particular are you referring to when you say that the market would discipline high leverage positions?


  44. Gravatar of Lars Christensen Lars Christensen
    30. August 2014 at 02:18


    I have an answer for you;-)


  45. Gravatar of ssumner ssumner
    30. August 2014 at 06:18

    Rajat, See my reply to BC.

    Ben, Japan should achieve at least 1 percent per capita (which would be zero percent overall.). China will slow considerably from 7 percent.

    Vivian, Haven’t had a chance to read that yet, but any change in wage stickiness will have very little impact on the macroeconomy. The stickiness of wages is . . . Quite sticky.

    Lars, If China is where Taiwan was in 1990 then you’ve proved my point. Multiply Taiwan’s GDP relative to the US in 1990, times China’s population, and they’d certainly be ahead of the US.

    The yen has appreciated in real terms since 1970, hasn’t it?

    James, I’ve traveled in east and western Germany, and also Switzerland, The PPP numbers (15 percent richer) seem about right to me.

    In any case, I know more about China, and it’s obvious to me that China’s price level is far below US levels.

  46. Gravatar of W. Peden W. Peden
    30. August 2014 at 07:24

    The Eurozone crisis presents an interesting dilemma for the British hardline conservatives. On the one hand, “The Euro is Bad” is the first principle of UKIP-type conservativism. On the other hand, Eurozone monetary conditions have been much less inflationary than in the UK. If tight money is so great, then didn’t the UK miss out by not being part of the Eurozone?

    To be fair, one hears plenty from some UKIP figures (Tim Congdon, and IIRC their leader Nigel Farage) about the benefits of expansionary monetary policy or (more euphemistically) “devaluation”. Farage has at least said that what Greece needs is a good dose of “devaluation”. So the actual leadership of UKIP may have a better understanding of monetary policy than their rank-and-file; Tim Congdon certainly does.

  47. Gravatar of Mikio Mikio
    30. August 2014 at 08:29


    I’m with Lars on this one, in the sense that China overtaking the US is certainly not a certainty.

    The main reason for my “hunch” is derived from the markets: not just recent currency moves, but also the consistent underperformance of financial assets vs. US since 2010, then Europe, and more recently even Japanese assets.

    And I am talking about long term trends, such as 40-week moving average prices, or even longer, over the past 20 years or so, not just fluctuations.

    In fact in total return terms its Italian government bonds that beat every other category in that asset class.

    It’s more evident in broad stock market prices.

    What I am saying is that markets have been consistently suggesting for several years that in RELATIVE terms, the prospects old G7 world is improving vs. the EM, including, China and Asia ex Japan. Only exception is India.

    So when the markets on a global level are consistently signalling something, I listen.

    I can imagine several things that could go wrong in China, and go right in the US (or EU and JPN, etc.), but at the end of the that would be speculative since it is something that might, or might not, happen in the future.

    To make a long story short: China and the EM are rising, but the relative shift in their favor maybe coming to an end.

    Lars offers one possible explanation or model simulation.

    Best M

  48. Gravatar of Tom Brown Tom Brown
    30. August 2014 at 09:47

    Jason, love the link, but you write:

    “but without having had a Great Depression or WWII earlier.”

    I’m sure you must mean that in some other sense (that the obvious one), since clearly China had more than it’s share of WWII. 😀

  49. Gravatar of Daniel Daniel
    30. August 2014 at 10:11

    Somewhat relevant – let’s hope China is better at playing great power games, because the US govt just sucks at them


    THIS is the real reason to oppose US imperialism – not for being an empire, but for being crappy at it.

  50. Gravatar of Sumners on Christensen on China « Economics Info Sumners on Christensen on China « Economics Info
    30. August 2014 at 22:00

    […] Source […]

  51. Gravatar of James in London James in London
    31. August 2014 at 04:00

    Simple ratios like total assets to tangible common equity, but also the market had an understanding of who owned the most high risk assets. It knew which banks held the biggest percentage of sub-prime mortgage loans or sub-prime HELOCS relative to total loans; which banks and others owned the biggest percentage of securitised debt based on such assets; which banks and others held derivatives based on indexes of such securities. Those derivatives have little or no market value at the time they are contracted, so leverage is very hard to spot sometimes. They can have huge negative positions, black swan-like tail risk, if certain events occur.

    As the tide went out the market was incredibly efficient at picking out those most exposed at any one time. Most financial players had to mark to market these positions, and often had to meet margin calls to cover the growing mark to market losses. Not readily meeting margin calls are huge red flags to the market, and traders and risk managers then run from those institutions, reinforcing the liquidity squeeze and ratcheting up the demand for money.

    However, at some point the tide had gone out so far that a panic built up, demand for money began to go exponential, aggregate demand dropped sharply and nominal GDP sagged. At that point, around the time Lehman was caught out, but probably earlier, the big central banks should have acted. They, and most of the smart people (including dumb me) were not looking at what orthodox monetary theory said they should have focused on. Dumb traders and risk managers were, though, they could feel the liquidity squeeze caused by the ever rising demand for money and the fact the central banks weren’t meeting it in any meaningful way.

    Events move very quickly at these times and institutions who didn’t appear leveraged suddenly realise they are as things they own start to drop in value. The market is brutal at such times. It became a panic because the markets weren’t seeing the lenders of last resort step up to the plate anything like quickly enough. It’s ironic to see the 2008-09 called a crisis, when really it was a panic, just like all the old ones. A “panic” implies central bank mismanagement, and so far the central banks have successfully avoided public responsibility – successfully pinning all the blame on the greedy (commercial) bankers.

  52. Gravatar of Kevin Erdmann Kevin Erdmann
    31. August 2014 at 09:59


    Thanks for the thorough review. But, one thing that I’ve noticed looking back on the episode is that, in many (but not all) ways, it seems like the data don’t match the narrative.

    For instance, banks generally continued to hold much of their real estate loans on their books. And, from the Federal Reserve H8 releases, Commercial Bank Total Assets were declining, relative to “Residual” equity. Am I reading that wrong, or is there a reason you think other data is more useful?

  53. Gravatar of E. Harding E. Harding
    31. August 2014 at 10:54

    Another exchange rate absurdity: under an exchange-rate basis, Russia has an economy the size of that of Italy and smaller than that of Brazil. Roughly nobody in any of these countries believes this is true on a PPP basis.

  54. Gravatar of Luis Pedro Coelho Luis Pedro Coelho
    31. August 2014 at 10:55

    W. Peden, Farange is often very good on monetary policy. His party may have lots of nasty racists, but on monetary policy, he is pretty good.

  55. Gravatar of James in London James in London
    31. August 2014 at 11:06

    Not sure what you are driving at. Banks adopt different strategies. Some take more risk, some less. Industry-wide stats aren’t that relevant. The riskier ones do better in the boom, worse in the bust. IndyMac failed early, and the hit taken by senior creditors there spread panic to its vulnerable bigger brothers WAMU and Countrywide Financial.

    The most aggressive Investment bank was well known to be Bear Stearns, who fell first. Second most leveraged was Lehmans. The Lehman failure triggered more panic and forced the next most levered ML into the arms of BoA, it threatened MS and started knocking on the door of GS too.

    Wachovia was acknowledged as a good bank until they made what many thought a screwy decision to buy Golden West. So Wachovia fell into the arms of Wells Fargo.

    The narrative, the sequence of failures, is the way to see who was most levered. But to look at industry statistics is to miss the actual action, even if industry stats show up as leverage falling as the crisis built. What else would you expect in a building panic?

    And then this misses the monetary irresponsibility of the central banks also at work, seemingly frozen by academically interesting but practically disastrous TBTF debates and fussing about headline inflation at a time when aggregate demand is collapsing.

  56. Gravatar of Kevin Erdmann Kevin Erdmann
    31. August 2014 at 13:01

    Interesting, James. So, are you saying that macro trends are better described as variance among market participants than as aggregate trends? If that’s the case, it could be useful. Do you have any sources of data to suggest that can inform this sort of interpretation?


  57. Gravatar of ssumner ssumner
    1. September 2014 at 10:32

    Mikio, China has already overtaken the US.

    The Chinese stock market is a lousy indicator of future Chinese growth, as it is dominated by SOEs.

  58. Gravatar of Mikio Mikio
    2. September 2014 at 05:15

    Scott – well, by some measures, yes.

    But in terms of relative performance, that’s why I am not resting just on the Chinese SOE-dominated stock market. It’s a general, broader, EM underperformance. Unless you believe that all EM stock markets are lousy indicators, your argument has a weakness.

    Best M

  59. Gravatar of James in London James in London
    2. September 2014 at 06:00

    Dispersion of stock market returns is not a bad proxy, as would have been CDS spreads for a while until the LOLR reappeared.

    Tricksy issue is survivorship bias. The bad banks and failures get taken out over time. “At the time” (without hindsight) they all seemed equal, more or less. A good starting point for all the names lost (not all bad) is here:

    The FDIC has a database too, of course. They are the main clean-up agency as long as no systemic failures need sorting out.

  60. Gravatar of am am
    2. September 2014 at 06:25

    This country isn’t messing about.

    Some one, who hadn’t been to China for about 10 years, visited again recently. The obvious growth in wealth was very apparent including less sign of poor people. The latter are reducing in numbers as higher wages obtain throughout the country enabling a better standard of living for more.

    I agree with the general trend of the post and would say they have just started. There is many a big ticket item not yet produced in China. If Boeing and Airbus have their world market share in 10-20 years I would be a bit surprised.

    In a previous post the thought was raised on the ability to save in comparison to the US middle or upper middle class. It is by no means confined to China but I suggest that developing countries or even people who are coming out of poverty have a greater tendency to save for the rainy day or other reasons because the rainy day happened very recently and still is a painful memory. If rural Africa was taken as an example people might be surprised at how much is hidden in the house.

  61. Gravatar of Kevin Erdmann Kevin Erdmann
    2. September 2014 at 09:28

    Thanks James.

    The CBOE has implied correlation indexes, and corrections are certainly associated with increases in correlation. But, in terms of using something like this as an indicator, do you have a systematic method? It seems like a lot of indicators that, while they definitely move systematically with certain market fluctuations, they don’t provide a clear buy/sell signal, like the CAPE signal.

  62. Gravatar of ssumner ssumner
    2. September 2014 at 13:52

    Mikio, Two problems with your argument:

    1. I made no claims about the other EMs, so their stocks have no bearing on my argument.

    2. the EMH says that stocks move for many reasons, and that stocks do not reflect relative GDP growth rates. To the extent that growth matters at all, it is unexpected growth rate changes that matter. Thus EMs might have gone from being expected to grow 3 percent faster than the US, to a position where they are expected to grow 2 percent faster.

  63. Gravatar of James in London James in London
    2. September 2014 at 23:23

    The only indicator is as a result of hardwork digging into what the banks get up to. Two banks with identical loan portfolios on Day 1 will have very different loan losses on Day 901 as a result of very different methods of monitoring the loans and handling problems. Long term historic track record can be a good guide, but that has to be over more than one cycle, and the cycles can be 10 to 20 years in duration. If you are looking for anything obvious and short term you will look in vain. The EMH rules.

  64. Gravatar of CLingle CLingle
    27. September 2014 at 08:46

    Did someone say, “extrapolation” … ?

    Of course, in the real world political conditions seldome collide with the fantasies of “us” economists … ?

Leave a Reply