Monday, May 18, 2009

A tale of two banking crises: Japan and Korea

Economics may be a “science” but it lacks controlled experiments.  Especially in macroeconomics you can’t repeat an experiment with one variable changed and see how the single variable changes the outcome.  Economists have lots of statistical tools to deal with this – but those make the discipline either incomprehensible or diabolically boring.  [Apologies to all those who taught me econometrics.]

But every now and again people throw up a controlled experiment – two situations that are very similar and differ markedly only in one major element.  Yet strangely these situations seem under-studied. 

What I want to do here is give a stylised version of Japanese and Korean economic history and how it pertains to the banking crisis both countries had.  My knowledge of this however comes the way much of my stuff comes – from the history of the banks backwards.  So I am sure to offend people with deep understandings of the political/economic history and I welcome someone telling me I am just wrong.

First however I need a stylised history of Japan starting with the arrival of Commodore Perry’s black ships in 1853.

Before Perry Japan was almost autarkic.  There was a relatively weak central government and about 300 “han” – being relatively strong feudally controlled districts.  The emperor did not effectively speak for Japan when Perry came in, guns blazing.

The Meiji Restoration changed this.  Japan was reformed as a centrally controlled empire – with a ruling oligarchy ruling through the Emperor who claimed dominion over all of Japan.  The “han” were combined to form (75?) prefectures with a governor appointed centrally.

The view of the new oligarchs was that Japan would get rich through (a) industrialisation and (b) unequal trade treaties to match the unequal treaties imposed on Japan by Perry et al.  To this end they invaded Korea and started the military industrialisation that ended eventually with World War 2.  There were major wars in Korea and against an expansionist Tsarist Russia (especially 1904-1905). 

Ok – that is your 143 word history of Japan from Perry to World War 2.  Like any 143 word history it will leave out important stuff.  I just want to focus on how this foreign policy adventurism was financed.

Financing Japanese expansionism - and that financial system until today 

Firstly it is simply not possible to expand heavy industrialisation of the type required by an early 20th Century military-industrial state without massive internal savings.  Those steel mills had to be funded.  And so they set up the infrastructure to do it. 

Central to this was a pattern of “educating” (the cynical might say brainwashing) young girls into believing that their life would be happy if they had considerable savings in the form of cash balances at the bank (or post office).  Japanese wives often save very hard – and are often insistent on it.  The people I know who have married Japanese women confirm this expectation survives to this day. 

Having saved at a bank (and for that matter also purchased life insurance from an insurance company loosely associated with the bank) the financial institutions had plenty of lendable funds.

The financial institutions by-and-large did not lend these funds to the household sector.  Indeed lending to the household sector was mostly discouraged and was the business of very seedy loan sharks.  To this day Japan has a relatively undeveloped credit card infrastructure with very high fees.  These high fees are a throwback to the unwillingness of the institutions to lend to households. 

Japanese banks instead lent to tied industry – particularly heavy industry.  It was steel mills, the companies that built power plants, the big machine tool makers.  Many of the companies exist today and include Fuji Heavy Industries, Kawasaki Heavy Industries and other giants such as Toshiba.  Most of these super-heavy industrials were tied to the banks (and vertically integrated) called Zaibatsu. 

Now steel is a commodity which has wild swings in its price.  Maybe not as ordinarily wild as the last five years – but still very large swings.  And these steel mills were highly indebted to their tied banks.  Which meant that they could go bust.

And as expected the Japanese authorities had a solution – which is they deliberately cartelized the steel industry and used the cartel (and import restrictions) to raise prices to a level sufficient to ensure the heavy industry in question could service its debt. 

The formula was thus (a) encourage huge levels of saving hence (b) allow for large debt funded heavy industrial growth.  To ensure it works financially (c) allow enough government intervention to ensure everyone’s solvency.

When the Americans occupied Japan their first agenda was to dismantle the Zaibatsu.  They were (in the words of Douglas McArthur) “the moneybags of militarism”.  

Like many post WW2 agendas that agenda was dumped in the Cold War.  The owners of the Zaibatsu were separated from their assets and some cross shareholdings were unwound – but the institution survived – and the Zaibatsu (now renamed Keiretsu) remained the central organising structure of Japan.  Dismantling Japan’s industrial structure did not make sense in the face of the Korean War.  The pre-war Zaibatsu had more concentrated ownership than post-war Keiretsu.

The point is that it was the similar structure before and after the war – and it allowed massive industrialisation twice – admittedly the second time for peaceful purposes. 

Now the system began to break down.  Firstly by 1985 steel was not the important industry that it had been in 1950 or 1920.  Indeed almost everywhere you looked heavy industry became less important relative to other industrialisation.  By the 1980s pretty well everywhere in the world tended to look on such heavy industries as “dinosaurs”.  This was a problem for Japanese banks because they had lent huge sums to these industries guaranteed by the willingness of the State to allow cartelisation.  You can’t successfully cartelise a collapsed industry.

Still the state was resourceful.  Originally (believe it or not) they opposed the formation of Sony – because they did not know how to cartelize a transistor industry.  Fifteen years later the UK Prime Minister French Prime Minister  President would refer to his Japanese counterpart as “that transistor salesman” and he was not using hyperbole.  Still the companies coming out of new Japan – technology driven mostly – did not require the capital that Japan had in plentiful supply.  If you look at the companies coming out of Kyoto (Japan’s Silicon Valley) they include such wonders as Nintendo – companies which supply huge deposits to banks – not demand huge funds from them.  [Incidentally in typical Japanese fashion the biggest shareholder in Nintendo is Bank of Kyoto.  Old habits re-cross shareholdings die hard.]

The banks however still had plenty of Yen, and they lent it where they were next most willing – to landholders.  The lending was legion and legendary – with golf clubs being the most famous example of excess.  [At one stage the listed exchange for golf club memberships had twice the market capitalisation of the entire Australian stock exchange.]

Another place of excessive lending was to people consolidating (or leveraging up) the property portfolios of department stores.  Think what Bill Ackman plans to do to Target being done to the entire country – and at very high starting valuations. 

Meanwhile the industrial companies became zombies.  I have attached 20 year balance sheets for a few of them here and here.  These companies had huge debts backed by dinosaur industry structures.  They looked like they would never repay their debts – but because they were so intertwined with the banks the banks never shut them down.  As long as interest rates stayed near zero the banks did not need to collect their money back from them.  As long as they made token payments they could be deemed to be current.  There was not even a cash drain at the banks at low rates.  The rapid improvement in the zombie-industrial balance sheets in the past five years was the massive boom in heavy industrial commodities (eg steel, parts for power stations etc).  Even the zombies could come alive again…  only to return to living dead status again quite rapidly with this recession.

Anyway – an aside here.  Real Japan watchers don’t refer to the banks as zombies.  They refer to the industrial companies as zombies.  (Although most of the Western blogosphere does.)  

Most of the banks had plenty of lendable funds and a willingness to lend them.  They did not have the customers – and the biggest, oldest and most venerable of Japanese companies were zombies.  So were the golf courses, department stores and other levered land holders.  I get really rather annoyed when people talk of zombie banks in Japan – it shows a lack of basic background in Nihon. 

Note how this crisis ended.

1).  The bank made lots of bad loans – firstly to heavy industrial companies and secondly to real estate related companies (golf courses, department stores etc). 

2).  The loans could not be repaid.

3).  The system was never short of funding because the Japanese housewives (the legendary Mrs Watanabe) saved and saved and saved – and the banks were thus awash with deposit funding.

4).  The savings of Mrs Watanabe went on – indeed continued to grow – with zero rates.

5).  Zero rates and vast excess funding at the banks made it unnecessary for the banks to call the property holders and (especially) the industrial giants to account for their borrowings.  Everything was just rolled.

6).  Employment in the industrial giants of Japan thus never shrank (Toshiba alone employs a quarter of a million people).  The economy continued to sink its productive labour force into dinosaur industries and dinosaur department store chains.

7).  The economy stagnated – but without collapse of any of the major banks and without huge subsidies to the banking system.  [The number of banks – mostly regional banks – that failed during the crisis was not large given the depth of the crisis.]

Now lets look at Korea. 

Korea was occupied by Japan until the end of WW2.  They chose to industrialise in the pattern they understood – a Japanese pattern.  For Keiretsu substitute Chaebol and you have the idea.  The Chaebol were private heavy industrial conglomerates tied to financial institutions and with intense government support. 

And the Chaebol suffered the same fate (slow irrelevance of heavy industry) as the Japanese heavy companies except they were called to account and many of them failed.

The reason is the different banking structure.  Korea started its Chaebol industrialisation later than Japan – and the one multi-generational part of the formula (educating young women that they should save and save and save) was just not done as well.  This is a multi-generational process.  

The result is that the Korean banks – unlike their Japanese counterparts – were short funds.  Endless funding at zero interest rates was simply not possible.  Given that the banks eventually collapsed – with many becoming government property and with the government winding up as the largest shareholder in almost all banks.  This was a spectacular crash – as opposed to a slow-burn malaise.  Chaebol failed.  In some instances their founders were imprisoned.  The strongest Chaebol is the one most associated with new industries (Samsung).  It survived and prospered – but others did not.

Korea had a much worse recession than Japan.  Vastly worse.  Japan was just low growth for a very long time.  By contrast the Korean economy crashed and burned.  But it also recovered very fast and at one point (1999-2000) the Korean Stock market was 1932 Great Depression cheap.  It bounced. 

It is my contention that the main difference between the Korean and Japanese crashes (and Korea’s case recoveries) was the funding of the banks.  In this view Korea’s was so sharp because the banks simply ran out of money – and that caused massive liquidations across the economy – systemic failures. 

The recovery was also sharp because the systemic failure meant that businesses that shouldn’t have failed (because they were profitable worthwhile businesses) got into deep distress.  Real companies died not because they deserved to die but because the system in crisis killed them.  There was a case for bailing out those companies – and the rapid recovery told you this was something systematic – not business specific.  The massive upward movement in the stock market at the end of the crisis was the secondary proof that good businesses were killed.  It was also probably the best investment opportunity globally in the last twenty years.

The economic decline in Japan was so gradual and so sustained precisely because there was no systemic failure and no reason to reallocate resources from bad businesses to good businesses.  Zombie companies could exist for decades – and there was no renewal.  A little bit of failure would have been a good thing – creative destruction.  And the survival of bad businesses in Japan is part of the reason the stock market never bounced there.  No investment opportunities.

Policy question:  how do you ensure the creative destruction without putting the good bits of the real economy to the sword? 

Investment question: what bits of the USA (and the rest of the world) will wind up looking like Korea and providing the best investment opportunity in two decades?  And what bits will look depressed for two decades before going into a bit of a decline? 

For discussion.  And thanks for bearing with a long post.



Eddie Bravo said...

John interesting post - would point out something though.

Zaibatsu and Keiretsu are distinct and separate concepts.

Zaibatsu means literally financial clique in Japanese. The clique in this case being the companies have a core bank that finances their operations.

Thus, any company with a Sumitomo, Mitsubishi, or Mitsui in its name is part of a zaibatsu. They can by virtue of their name rely on their respective bank - Sumitomo Mitsui Financial Group, or Mitsubishi UFJ Financial Group - to furnish them with easy financing.

Hence, the continued survival of a ropey automaker, such as Mitsubishi Motors, or shipping line, Mitsui OSK.

Keiretsu merely means linked system/thread (I am taking the literal meanings of the Japanese characters used for both zaibatsu and keiretsu here). While there will be a central company at the heart of the keiretsu these are not banks.

Auto companies are the classic keiretsu - Toyota keiretsu etc. Toyota Motor is the central company and there are litereally hundreds of affiliated parts suppliers and body assembling companies that are legally distinct entities but rely on anywhere from a sizeable chunk to almost their entire revenue on the central keiretsu company. Like the zaibatsu group bank the central keiretsu firm will almost certainly have an equity stake in their affiliates but the keiretsu firms cannot rely on easy low-cost funding initiatives that the zaibatsu group bank would roll out to their firms.

Thus small keiretsu firms will likely go to the wall in real crisis moments. Have a look at the small keiretsu companies failing in Nagoya (toyota land) - granted these aren't the biggies like Denso but almost all the economy in nagoya is geared to Toyota and small firms will go under. In this sense, keiretsu is a lot more diffuse than Zaibatsu.

In addition, zaibatsu firms that are generally a lot bigger in scale than the multitude of keiretsu firms. Like Toyota and Denso, due to the size of most Zaibatsu firms, the bank will not let these fail or will try and cook up a way for it and other zaibatsu family names to bail out the straggler.

Keiretsu would be more analogous to what you see at GM, GMAC and Delphi, while Zaibatsu is more similar to what you see at GE in terms of a pure conglomerate with a financial institution rolled in. Although unlike GE the group companies are also listed separately but linked via cross-shareholdings. I.e you can buy pure equity exposure to Mitsubishi Heavy Industries but not to GE's turbine business.

Finally, regardless of Korean women's propensity to save, Korean chaebol never had a group bank - you did not have "Samsung Bank" or "Hyundai Bank" in the way you have SMFG or MUFG in Japan. or their precursors - Mitsubishi Bank, Sumitomo Bank and Mitsui Bank.

This I think is a big factor in explaining how Hyundai suddenly cratered while say Mitsui Bussan did not. Mitsui Bussan always has a friendly bank that is prepared to waive them some time (that can be indefinite if need be). If there was a Hyundai bank then this would have been probably quite different for Hyundai.

Also, I think this is one of the reasons that Nissan (a keiretsu rather than a zaibatsu) needed to be bailed out by Renault - there is no Nissan Bank to help it out, in contrast, to Mitsubishi Motors. And why say private equity funds have been able to buy into auto-parts makers (keiretsu firms) in a way that is impossible for them to buy into a zaibatsu firm.

PlanMaestro said...

No John, thank you for a great post. Worth every word.

If you have an opinion on which sectors will be great buying opportunities like Korea, or stay at depression levels for ever like Japan, please share. I imagine that with your mention of Nintendo and Samsung, versus the over-levered real estate, retailing, and heavy industries you gave us a hint.


John Hempton said...

I acknowledge the shorthand - the Zaibatsu had common ownerhsip - the Keiretsu only a meeting on the third thursday of the month (as per Mitsubishi).

Only the ones organised around banks really count.


Anonymous said...

re investment implications. The well capitalized banks (JPM, many regionals that passed on the subprime boom) and new financial institutions that spring up will be good investment candidates as there has never been a better time to become a bank in the US for at least the last generation (steep yield curve, low cost of funds, plethora of investment opportunity, still $40 trillion of household wealth to intermediate).

Also think stick with what US is good at; high tech as the sector has outperformed throughout crisis and US spends half of global R&D and tech good at competing in deflationary pricing regimes. And healthcare looks very cheap historically and on valuation metrics but obviously could very well be a value trap as it's been the last 5 years and obviously lots of policy risk.

sudhee said...

1. Medical and Healthcare services within US.
2. Energy and energy efficiency companies.
3. Export driven companies from the US (as $ revalues against major currencies)

Anonymous said...

Interesting post. Seems like one clear lesson is that you can only follow the Japanese path (leaving aside for the moment the question of whether this path is desirable or not, which I understand is an issue of great controversy) IF you have the savings to do so.

So the US can continue to emulate Japan only if China (the equivalent of the Japanese housewife) continues to save and provide effectively zero-rate financing.

Anonymous said...

"how do you ensure the creative destruction without putting the good bits of the real economy to the sword?"

You have to acknowledge that creative destruction is imperfect and some good bits will fail. To use an unpleasant, but honest analogy: There is guaranteed to be some collateral damage. Avoiding collateral damage to the point the you refuse to fight the war is called appeasement -- and that is unlikely to bring us to a good end.

himaginary said...

Very interesting post.

But just one minor point I noticed. The famous transistor salesman phrase was from French president Charles De Gaulle.

John Hempton said...

I write this - mentioning how difficult Toshiba's balance sheet is - and lo - one evening later

The book for the Japanese Domestic offering for TOSHIBA (6502 JT) is now open.

* Books open: 18 May
* Books close: X - 1 day at 5pm (London time) for international investors; X – 1 day at 11am (Tokyo time) for Japanese Domestic Investors;
* Pricing and allocation: X (one business day between 27 May – 2 June)
* Payment and delivery for International investors: X + 6

Order Taking
* Discount range: 3% - 5% to the closing price as of the Pricing Date, they may place limits at a discount of either 3%, 4% or 5% to the closing price on the Pricing Date.
* Books open on 18 May
* Orders for international investors can be taken by Nomura only.
* Investor Letter: International investors are required to sign and return the Investor Letter. No allocations to the accounts who have not returned it before book closes.

Unknown said...

Great post.

I think a commentator above hit it right when pointing out that creative destruction is not perfect.

And I think you have to remember that bankruptcy usually results in the assets being re-allocated. The best parts of the good firms are often again put to use elsewhere.

Geoff said...

Great post, and very interesting comments by Eddie Bravo. One point I would make is not to discount the role of the big trading companies (Mitsubishi Corp, Mitsui Corp, Sumitomo Corp, Itochu and Marubeni) as a source of capital in Japan now.

Their role in providing both equity and debt capital to Japanese (& foreign) corporates should not be underestimated, and they are some of the few institutions in Japan to really understand the cost of capital and only to deploy funds where the projected returns exceed the WACC.

These are the reasons like Mit Corp has a five year rolling RoE > 15% whereas most of Japan Inc makes low single digit RoE's.

Anonymous said...

The Japanese prime minister was Hayato Ikeda, and the head of state was not the British PM at all, but Charles de Gaulle:

William K. Tabb, The postwar Japanese system, p.14

John Hempton said...

French Prime Minister is corrected...



Anonymous said...

(Bangs head against wall)


Charles de Gaulle was NOT the prime minister of France! He was the first President of the 4th and 5th Republics, and the remark was made in Nov 1962.

In the 5th Republic, most executive authority is vested in the President. The premier is more like the US White House Chief of Staff, except that she must meet with the approval of the National Assembly.

Unknown said...

John, Thank you for this fascinating post. Could you just clarify a little further as to why the stars were so aligned in Korea post-crisis- marginal companies got wiped out, but you say decent ones did too. Are you saying the real economy wasn't in as bad shape as the financial industry and the banks? How did the wise investor avoid those decent companies that got harvested by the crisis? Thank you again, this was such a great read.

Anonymous said...

Seems to me the difference between this crisis and every other crisis : corporate lending and commercial real estate, this time household debt & mortgages. This adds a political dimension.

John Hempton said...

President/Prime Minister - fixed again - and sorry.

My French history is weaker than my Asian history which reflects my location and interests.


AndyG said...

But whats the difference between Japanese housewives save save saving and the huge holdings of US Treasuries by China & Japan?

JessicaYogini said...

Thank you for the interesting post. You have made me rethink things.
I had thought that the East Asian model of funneling savings away form consumption and into investment and dealing with the resulting overproduction through exports was a catch-up system that can not deal with the transition to caught-up. And that Japan has done worse than Korea (and Taiwan) because it was simply farther along the curve.
Since China is using a similar model, it will be a big difference whether they can or can not shift away from export dependence.
I have been hearing about "naijukakudai" (expansion of domestic demand) for nearly a quarter century now but I have not seen any of the nations that exported themselves into the 1st world succeed in doing it.

L.E. said...


I enjoyed reading your article, a tale of two banking crises.
But I wanted to point out one thing about Korea.

So you mentioned that unlike Japan, Korea was short of funding because there was not enough saving by wives or whoever else.

Well, first of all, China, Japan, and Korea, as Confucian societies, has regarded frugality as best practice. Hence, even without government intervention, people tend to save rathern than to spend (this all changed now). Korea might haven't had savings rate as high as Japan, yet its development model has been noted with high savings rate. So, when you said that Korea banks did not have money to finance chabols, you sorta nuanced that it was because there was not enough deposits within Korean banks.

This argument then does not account what was happening within chabols and banks during that time.
Banks had money, yet didn't have enough money to fund all these manufacturing projects that were rapidly expanding by chabols in the early 90s. Hence, these projects were financed by international bank loans. Haggard and Mo (i believe 2001) argue that chabols were expanding investments in heavy-industry manufacturing projects 1)yen was rising against dollar, so wanted to take japanese market share aways, 2) korea was getting out of the 2nd adjustment period, and 3) chabols wanted to move from labor-intensive industry to capital-intensive as labor costs were rise rapidly. At the same time, the newly elected Kim Young Sam administration employed looser monetary policy to promote rapid economic growth in Korea. Also, various financial liberalization policies were adopted, hoping to join the OECD. Then, Cho(1998) argues that chabols expanded these projects even more, because once Korea join the OECD, cheaper international money would be available. Hence, in Korea, it wasn't really about shortage of funds within banks, it was more about over-expansion by chabols exceeding banks' deposits and about cheap money flowing in that financed chabols' projects and consequently made them vulnerable when capital flight occured.

John Hempton said...

The last comment is almost spot on. Korea had a high savings rate. Not Chinese - but certainly Asian. They had a VERY HIGH investment rate.

The industrialisation of Korea was faster (with the result that the South looks like an industrial theme park).

But the Korean banks remain modestly short deposits. After the Korean crisis they were long deposits again (with a huge crash to get there). But when I last had a really decent look at Korean banks (2006) they were again gearing up a deposit war.

A price war for deposits is unthinkable in Japan. Its reality in Korea - and that tells you that shortage of deposits - caused by savings being too low relative to investment - is a feature of the system.


PS. The Korean academic who wrote the last comment. Thanks.


L.E. said...

Moody's just downgraded the korean banks' BFSR, BCA, and GLC.

So, yes you are right about deposit short happening among korean banks.

Deteriorating balance sheets of the Korean banks and extreme rally in the Kospi are actually worrisom
[this could be a great opportunity for investment]

John Hempton said...

I am not sure that controlled experiments are necessary for science - but some kind of process for reasonably accepting or rejecting theories is...

Evolutionary biology also lacks controlled experiments - but genetics provides some pretty good tests that were not available when Darwin wrote his treatise.


Anonymous said...

but can the US Fed just print savings as required?

and if you koreans don't like it them we will cause some major collatoral damage in your towns.

Look what we did to Iraq after the terrorists decided to stop selling oil in US dollars.

Alexandra said...

That's an interesting article. You are in fact comparing different brands of capitalism, the Japanes, the South Korean and the American, neither of which seem to be particularly sucessful.

The Japanese obviously thought that keeping people at work was the important factor to keep the economy going - working people consume and save. That kept Japan running.
It seems that the Japanes banks were funneling the money to industrial enterprises regardless of profits for themselves.

That attitude does not seem to work in the U.S. The Wall St. banks are completly disconnected from the world around.
There businessmodell is making money from money and know that they have access to the dollar printing press, that can be achieved regardless of whether the world around them fails.

A more extensive treatment of your question is here:

coolhwhip said...

Great article, very exciting reading

Parag said...

Thanks for this version of Japan and Korean economic history.
Now that the worst calamity has hit Japan recently, the economy is going to suffer like anything. Not only Japan for that matter, countries that closely tied up with Japan regarding various policies are going to take a set back.
I pray to God for the people of Japan to give them strength to face this situation.
Korea credit card crisis

General disclaimer

The content contained in this blog represents the opinions of Mr. Hempton. You should assume Mr. Hempton and his affiliates have positions in the securities discussed in this blog, and such beneficial ownership can create a conflict of interest regarding the objectivity of this blog. Statements in the blog are not guarantees of future performance and are subject to certain risks, uncertainties and other factors. Certain information in this blog concerning economic trends and performance is based on or derived from information provided by third-party sources. Mr. Hempton does not guarantee the accuracy of such information and has not independently verified the accuracy or completeness of such information or the assumptions on which such information is based. Such information may change after it is posted and Mr. Hempton is not obligated to, and may not, update it. The commentary in this blog in no way constitutes a solicitation of business, an offer of a security or a solicitation to purchase a security, or investment advice. In fact, it should not be relied upon in making investment decisions, ever. It is intended solely for the entertainment of the reader, and the author. In particular this blog is not directed for investment purposes at US Persons.