Saturday, July 12, 2008

Deflation and bank bailouts in Japan

Given what is happening with Fannie Mae at the moment I should share a little of the history of non-US bank bail outs. I will start with Japan and later do Scandinavia.

Japan was an unusual bank collapse. It happened despite excess savings in the system. This is really strange. Most bank collapses happen when there is a lending binge that drives excess or investment or consumption and with a current account deficit. (See for instance Korea – where there was excess investment or Argentina where there was excess consumption.)

Japanese banks always had (at least collectively) sufficient deposits. [See my post on 77 Bank to see just how much excess deposits they now have.]

But the Japanese banks lent very badly indeed. Part of this lending was to the "Zombie companies" but most was on property. The formalised golf-club membership exchange in Japan at one stage was worth a good multiple of the entire Australian stock exchange (including giants such as BHP and Conzinc Rio Australia). Golf club memberships were of course a pure-play method of speculating on land.

But you need to notice that the Japanese bank collapse looked very different from what is going on in America now. The Japanese bank collapse was not a collapse of funding – it was a collapse of asset values and solvency. [Exceptions noted.]

American financial institutions are now having wholesale funding runs (or finding wholesale funding is unavailable which amounts to the same thing). Japanese financial institutions did not need wholesale funding (most had deposits) and hence by-and-large did not have runs. [There were some institutions such as the long-term-credit banks and similar institutions which had wholesale funding – they were effectively nationalised.]

Many Japanese regional banks (Nishi Nippon for example) were breathtakingly insolvent at the height of the crisis but they remained liquid because they had plenty of deposits. Because they remained liquid they never actually failed.

The zero interest rate policy

Insolvent but liquid banks are the key to understanding Japanese interest rate policy. There are several prominent macroeconomists in America (led notably by Krugman but joined by Bernanke) who argue that the zero interest rate policy was insufficiently expansionary – and that monetary policy should have been eased until it induced inflation. In theory this could be done by flying a helicopter over Tokyo throwing out freshly printed 5 thousand yen notes. Indeed it was in a speech about Japan that Bernanke uttered the famous helicopter line.

The BOJ always thought this policy was “risky”. Krugman’s response was that it was less risky that the endless government deficits Japan ran. Krugman missed the point – the question was who was inflation risky for? My answer: the banks.

A hypothetical insolvent bank

Imagine a hypothetical insolvent bank. Suppose the bank has 90 in funding, 10 in “stated equity” and “stated” 100 in assets. [I have left the currency blank because this could be 100s of billions of yen or billions of dollars.]

And suppose that the assets are not really 100 but 70 good and 30 bad - and everyone knows about the bad assets.

Then the bank “really” has 70 in assets, 90 in funding and minus 20 in equity. This is a realistic picture of insolvent Japan in 1994.

If the bank was in a current account deficit country like America, Australia, New Zealand or the UK there would be an immediate problem. In a wholesale funded market the 90 in funding would be say 60 of deposits and 30 of wholesale funds – and the wholesale funding would leave. The bank would go insolvent quite rapidly (see Northern Rock which was very reliant on wholesale funding).

But in Japan the 90 in funding was all deposits and was sticky. The funding never left and the bank continued quite nicely. Capital market discipline was not imposed – and the hypothetical bank could pretend there was no problem for many years. Banks fold when they go illiquid - not when they go insolvent. No liquidity problem means no crisis.

But the problem is still real. Over time insolvency may turn into illiquidity.

Now suppose that (and this is a gross simplification) that the spread between deposit rates was 2%. But rates could either be 10 and 12 percent or 0 and 2 percent.

If the rates were 10% and 12% the 90 of funding would cost 9 per year. The 70 of “real” assets would yield 8.4 per year. The bank would be cash flow negative. Anything that is cash flow negative for long enough goes illiquid eventually. The insolvency problem would turn into a liquidity problem.

Now suppose the rates were 0% and 2%. The 90 in funding is free. The 70 in assets yields 1.4%. The same banks is cash flow positive in a low interest rate environment. If they are cash flow positive for 15 years the bank will fully recapitalise.

  • Summary: zero interest rates were critical to bank recapitalisation in Japan.

The reason why the BOJ rejected the Krugman/Bernanke line was that it was risky to the banks and the BOJ (and the MOF) are totally captured by their bank constituency. It was risky not to the economy but to banks. [Note the practice of amakudari translated as “descent from heaven” where former government officials get to be CEO of banks late in their career.]

Why this form of bailout won’t happen in America

In America it is the wholesale funded institutions that are in the most trouble. Think Bear Stearns, Lehman, Fannie, Freddie.

They are in diabolical trouble.

The funding is leaving them. It does not matter whether rates are 0 or 10 – the funding is still going.

America will look far more like Scandinavia than Japan. Scandinavia was a funding crisis. The posts by Naked Capitalism and others suggesting that Japan’s wasted decade will be the new normal are just plan wrong.


I still have not worked out what side of the inflation/deflation divide I am. But the people that point to Japan as a likely outcome miss a point. Japan chose deflation because the alternative was nationalising the banks.

America does not have that choice. The American institutions are wholesale funded and hence will nationalised or fail if the wholesale funding disappears.

Nationalisation can be inflationary if it involves printing. The date the Federal Reserve is not printing – but Helicopter Ben has made clear that in Japan the BOJ should have printed. And the institutional imperative to stop him printing will not be present in America.


Timothy said...

Hi John, you have reached a similar conclusion to me regarding Japan. I would add that Japanese voters were almost all savers so there was no constituency for inflation. In countries such as the US or UK were many voters are big debtors the temptation to try to inflate away the debt may be strong.

John Hempton said...

I had never thought of your voting extension but it fits the facts. The LDP is based in the non Tokyo areas of Japan and the savings rate is most intense in the non-mega-city areas of Japan.

There is no constituency amongst the areas where the LDP is strongest for inflation. Deflation is fine by them...

Scott said...

Absolutely outstanding piece...I agree 100% with your conclusions regarding Japanese banks and monetary policy.

bio-anomoly said...

If you are going to judge inflation/deflation by who has a vested interest, then consider that Japan was not in debt to the Arabs and China for $N Trillion.

Chinese and Arabs have most of their USA holdings in cash, and they would object to inflation. The Chinese have long had a problem that they don't even dare float the Yuan, for fear of losing their foreign reserve value.

If the USA deliberately caused inflation, there would be "capital flight".

John Haskell said...

@bio-anomoly (sic)

Remind me again, how many electoral votes does China have? Oh, zero, you say? So the loudest objectors to inflation have absolutely no political voice in America, then?

Well I guess you know now which way this thing is going.

Anonymous said...

I was doing some research on google about japan and bank nationalization and came across this post and have been hooked on your blog all evening.

Really - your blog is a treasure. Just wanted to post a thank you.


variantperceptions said...

John, just to say that for a new blogger to reach conclusions that you reached two years ago is very discouraging. I hate you :)

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.