Thursday, January 22, 2009

Zero in Japan versus zero in America

I don’t have a solid economic model of different types of zero interest rate policies. I guess the number of instances historically doesn’t give us enough material to get a classification. But this is worth stating.


Zero in Japan looks very different from how zero in America looks right now.


How Japan looks


One of the first posts on my blog was about 77 Bank. 77 is a typical mid sized Japanese regional bank. It has vast excess deposits. It has plenty of liquidity – but almost nobody to lend it to. Competition to attract worthy borrowers is intense – and is led by prices. Loans are typically made on less than 50bps of margin.


The banks are bizarrely unprofitable. At 12 times leverage and a spread after costs of about a third of a percent the return on equity – before tax and provisions – is about four percent.


Businesses – at least ones with any claim to be credit worthy – have no trouble borrowing at all in Japan.


There is however a problem with these low spreads – a very big problem. Japanese banks are vulnerable to very low levels of credit losses. A loss rate of a third of one percent wipes out profit. A loss rate of 2 percent sustained over a couple of years would wipe out half the capital.


Summary: Japan has zero rates and the banks can’t make a spread because they have no willing borrowers. The banks have no liquidity problems – but they have a capital problem whenever losses rise to merely low levels from extremely low levels. The reason banks stopped lending was that there were insufficient willing borrowers.


The broken bank in Japan is a “zombie” (living dead). It has insufficient capital and not enough spread to rebuild capital over even a decade. It has enough liquidity to limp on for many years. However it often is attracted to riskier loans in a vain effort to find some spread – and it is prone to blow ups. For a long time Nishi Nippon City bank looked like that.  


Note that Japan is now facing credit losses of a couple of percent - and that is high enough to cause widespread devastation to the low-spread Japanese banking system.  


How the US looks


Most US banks have spreads after costs well above two percent of assets. Wells Fargo for a long time has had an interest spread above five percent. Spreads are still that high or higher.


In the good times background loss rates were one percent or more. Indeed there were plenty of businesses in the US which worked fine with three percent background loss rates (much car lending for instance).


There is no shortage of willing borrowers. Indeed there are plenty of worthwhile projects at the moment that have a hard time being funded because the banks don’t have any money available. Banks can’t fund themselves as the market for senior bank funding has shut down. The reason banks have stopped lending is that there are insufficient people willing to provide funds to banks. In plan parlance banks don’t lend because they can’t borrow.


Some observations


It is trite – but the key difference is between a current account deficit country and a current account surplus country.


The deficit country has to borrow from abroad. Banks intermediate the deficit and the banks are subject to hot money flows.


Being subject to hot money flows the banks are subject to runs – very big and destructive runs.


The banks in those countries fail fast. Japanese banks by contrast remain as zombies (living dead) with insufficient capital but enough liquidity to last decades.


In current account deficit countries nationalisation or part nationalisation is a common end-game for a financial crisis.  


Lots of people talk about Sweden (or better Norway) because they did their nationalisation well. But another example is Korea – where the bank blew themselves up too – and were critically dependent on (Japanese) money which became rapidly unavailable.


Help please


I know how these things look. I don’t have a decent model grounded in facts-on-the-ground.


I am surprised at the tone of the WSJ story about Japanese regional banks needing a bailout. They don’t need it from a liquidity perspective – they need it from a capital perspective. The WSJ story does not make this clear.


In America and the UK the banks have (serious) liquidity problems. I am not sure they have capital problems. Indeed my view is that there is no capital problem in the system – but the banks individually might have issues. The reason there is no capital problem in the system is that the underlying pre-tax pre-provision profitability is about 400 billion per annum.


Am I right that the system has adequate capital?


This view looks really controversial. It is just assumed – more or less by all pundits – that the banks are insolvent. Krugman’s latest piece on zombie banks is just one of many.


But the pre-crisis net tangible capital of the US banks was about 1.4 trillion. About 500 billion has been raised or defaulted since the crisis began. So call it 1.9 trillion. The pre-tax, pre-provision profitability has added another 400 billion per year to the pool – so we are at 2.3 trillion or more. A few hundred billion of the losses are borne outside the banking system.


If total losses get to the 3 trillion numbers that Roubini talks about. the system will get to neutral capital in two years and be fully recapitalised in five. If those numbers are right this is not a capital problem – it’s a liquidity problem.


The problem is serious though – and nationalisation may be the right prescription. It appears – as a matter of fact – to be the end game in countries which have current account deficits and banking crises (see Korea, Sweden, Norway). I am just not sure how to do it right – and when people (like Krugman) borrow the expression “zombie banks” from Japan I think they are substituting words for clear thinking.


I guess the thinking is hard though. If I thought all this stuff through clearly enough I would have that Nobel Prize in economics. Alas I did my last academic economics twenty years ago.






John Hempton



PS.  UK banks have (a) lower spreads and (b) lower starting capital.  The problems in the UK are thus more serious even with a lower level of losses than the US.  I am not sure that Barclays can ever be made solvent.  Barclays by comparison think they are solvent now.  But then I am a noted doubter of Barclays as one of my early posts show.  

PPS.  I was a little quick when I described Nishi Nippon City bank.  It is in fact a merger of two banks (Nishi Nippon and Fukuoka City bank).  Both were zombies.

12 comments:

cap vandal said...

"The reason there is no capital problem in the system is that the underlying pre-tax pre-provision profitability is about 400 billion per annum."

This would work fine if the banks could book exactly $100 billion per quarter in writedowns.

BAC and C ate up that much by themselves in 1Q. Maybe 1/2 of it, including the MER piece of BAC.

Anonymous said...

Josh- Excellent work on this blog, thank you for sharing your insights.

You have mentioned your disdain for Swedbank and its Baltic/ Ukrainian adventures in the past. Does not the entire Swedish banking system suffer from the same short liquidity position as English or U.S. banks?

By my calculations the Swedish banks have a net FX liability of more than $120 billion to the rest of the world, approaching 40% of Swedish GDP. And what of the Swiss banks?

Shouldn't banks in small economies like Sweden or Switzerland be "under the harrow" where the banking system is large relative to GDP and most of the funding comes from foreign sources? And if so, given all the discussion of the last Swedish bank bailout, when should we expect the next bailout/ nationalization of Swedbank itself?

John Hempton said...

Cap Vandal is right. In Japan where there was no liquidity problem the banks did leak their losses out over a decade.

They could because they had sufficient liquidity.

That is not the case in America. American (and UK banks) die with a bang - not turn into zombies.

J

John said...

I enjoy your insightful comments and occasionally provocative writing style. I was curious if you have ever taken a look at banks in Brasil and their current valuations. Good investment?

Mark Cheshire said...

I continue to be baffled at how Spain continues to delay its day of reckoning. The explanation is that the Bank of Spain enforced very conservative, counter-cyclical provisioning. However is that really enough, stacked up against the 10% current account deficit and the massive defaults in the construction industry? The conservative provisioning regulations did not exactly hold back growth in the banking sector which went like gangbusters, and it has to be one of the most over-covered retail sectors with a branch on every block in most citys. The shares have taken a hit, but there is no sense of crisis on a level with other countries. If people see weak points it is only among the smaller cajas and not the big banks. In fact the relative situation is such that the banks are complaining of unfair competition with other Euro-banks that have government support.

Any thoughts on why the bullet has been dodged so long? How much longer before crisis overwhelms the system, or will provisions see them through the cycle?

tbrown said...

Your scenario on banks gradually earning equity back to get back to a level of capital adequacy only works if the banks liabilities are of a long enough duration.

If a bank had negative equity but could credibly earn its way back to solvency, the bond holders, wholesale funders, and depositors wouldn't hang around - what's in it for them?

They'd all demand repayment at the first available oppotunity and the bank wouldn't be able to fund its assets any more.

Alternatively, there's every chance that the regulator would call the bank in and demand that it gets its solvency back up to a regulatory minimum immediately. That means hugely dilutive rights issues and/or nationalisation.

While your maths is as usual impeccable, I just don't see banks being allowed to do this in the real world.

Thomas

John Hempton said...

I am wondering if that is the Tom Brown from Bankstocks.com - the old analyst I knew well (and disagreed with often).

The issue is not DURATION of liabilities. It is sufficient guarantee of payment. Financial institutions since time-immemorial have got solvent this way.

The Tom Brown I am thinking of played one once - it was called Metris. Insolvent for all money. But it came back.

If the governmnent backing was credible the companies could roll their debt.

Hell - I reckon they could have rolled their debt pre-Lehman/Wamu because there was not enough fear.

WaMu was the super-destructive one because it wiped out bank senior creditors.

Anyway - will you let me know if it that Tom Brown?

Thanks.

j

Anonymous said...

John:

How does this entry (http://businomics.typepad.com/businomics_blog/2009/01/japans-depression-and-zombie-lending.html) from Bill Conerly (Businomics blog) square with your take?

If I read him right, he's endorsing the quoted study's view that the Japanese "zombie" bank crowded out investment that would have gone into healthy banks. He also states that continued access to credit is what kept these "zombie" banks living dead.

Thanks, and great blog:

z.

anoninca said...

I just don't get why you think that any bank that relied on wholesale markets for funding should be supported by the government, while it tries to get its funding in order (except in the case that this would actually mean a collapse of the banking system). In the US we have thousands of banks that did not rely on wholesale funding. Why should banks with a crappy business model be subsidized?

We should just follow normal procedures, let the FDIC take over banks that effed up their funding structure and sell of the remnants (after wiping out shareholders and a big chunk of debtholders) to banks that didn't eff up. I can understand the government financing the purchases that are made by solvent bank, I can't understand subsidizing a bad business model.

Your blog posts generally assume there is nothing wrong with the wholesale funding business model. It seems to me that events have rather dramatically proved you wrong.

Anonymous said...

John, your argument seems to imply that capital is not necessary (which implies infinite leverage). Or have I misread your argument?

aspirin said...

is the underlying profit 400 b per year or is this your impression based on banks as they existed before with their accounting practices as before? how much of that profit was non-zero-sum financial engineering accounting?

would love to hear you write about how comp is done in banking someday. clearly a problem in this crisis that is acknowledged but not talked about much.

Anonymous said...

Josh, I think you are enumerating some important distinctions between the Japanese and US banking systems and the economic environment they find themselves in. In a PS you were also kind enough to note further differences with UK banks ... any chance that you could similarly comment on Australian banks and the economic environment that they are operating in?

General disclaimer

The content contained in this blog represents the opinions of Mr. Hempton. Mr. Hempton may hold either long or short positions in securities of various companies discussed in the blog based upon Mr. Hempton's recommendations. The commentary in this blog in no way constitutes a solicitation of business 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.