Global Economics

Economist Draws Parallels Between U.S., Japan Banking Crises

The main lesson from 1990s Japan: A bailout of banks is politically unpopular but ultimately necessary to save the economy

Richard Koo, chief economist at the Nomura Research Institute in Tokyo, discusses the grim parallels between the US and Japan, including the need for unconditional capital injections.

Do you see any parallels between the US now, and Japan in the 1990s?

It's looking like a replay of many of the events that happened in Japan. Take the example a few weeks ago, during the G-7 meeting in Tokyo, when the Japanese finance minister advised the US treasury secretary to carry out immediate capital injections to the US banks. Paulson couldn't commit himself—similar to Japan in the 1990s, when public opinion was very strongly against bailing out highly-paid bankers and irresponsible banks. Recall also that in 1992, then-Prime Minister Miyazawa wanted to help the banking system with public funds, but was slammed by the public. As a result, it became taboo for the government to talk about bailing out the banks for the next six years.

This being an election year in the US, something similar is happening. Everyone is talking about a borrower rescue, not a lender rescue. In Japan, the public's perception towards the banks only changed when the Japanese credit crunch hit home. And that was in 1997, many years after the bubble had burst. That's because Japan was different to the US in one important respect—Japan, until 1997, had funds to lend but nobody wanted to borrow. The Japanese corporate sector was rebuilding its balance sheets. Until 1997, the average Japanese did not feel much pain. If he wanted to take out a loan, the banks were falling over themselves to make him one. But after 1997, the combination of a weak yen and a sagging stock market hit the banks. They realised that they were no longer meeting the 8% capital ratio mandated by the Bank for International Settlements, and desperately needed to reduce lending.

To make things worse, then-Prime Minister Ryutaro Hashimoto was trying to reduce the government fiscal deficit—at the most inopportune time. It was clear that the economy was kept going only by the government fiscal stimulus, yet here was Hashimoto talking of raising the consumption tax and other tightening measures. As soon as he came out with those ideas in January 1997, the stock market collapsed and everybody sold yen. That really hurt the capital cushion of the Japanese banks and they had to cut lending.

So the Asian financial crisis of 1997 affected Japan?

Yes, even though, in a way, the Asian crisis was actually caused by Japan. While the yen was very strong against the dollar, in the 80-90 range, having factories in Southeast Asia made sense. But when the yen weakened to 140, it made no sense to have those factories, so the Japanese started closing them down. But as the Japanese were pulling out, Western money was pouring in, even into countries which were no longer competitive. They were becoming uncompetitive because they were linked to the US dollar, which was gaining strongly in those years and was amongst the strongest currencies in the world. But the yen was going the other way. So the Southeast Asian business model was not sustainable, and not surprisingly, the dollar pegs were attacked by the likes of George Soros.

What was the next step in Japan after the credit crunch you just mentioned?

After the credit crunch, there were finally capital injections in March of 1998 and 1999. The second capital injection basically ended the crunch. This is what we have to see happen in the US. The problem is there are 8,500 banks in the US (there used to be more). These local banks have no access to sovereign wealth funds or other bailout money—if the big banks were ok, then they could help, but the big banks are not ok and Warren Buffett can't solve the banking problem on his own! So the small banks have no choice but to reduce their lending. And that will create a major credit crunch in the US.

What do you think will the US public's reaction to a bailout?

This is the problem. The Japanese government listened to the US advisers from the Treasury, who insisted on 'conditionality.' This essentially meant forcing the banks to make a million changes before they could get the money from the government. But this was utter nonsense. The Japanese banks simply refused to accept the government money. Instead, to get themselves back to the required capital asset ratios, they just stopped lending. But that's catastrophic for the wider economy.

So there should be no or just minimal conditionality in the US. These measures are to save the country, not the banks. In a situation like this, you have no choice. In Japan, the attempt to force conditionality onto the banks was abandoned because not a single bank applied for capital injection. The subsequent injection finally ended the credit crunch.

The length of time before the capital injection is implemented depends on the pain threshold of the US taxpayer. When the ordinary citizen, the tax payer, starts suffering serious pain because he can't get a loan from his bank, then he will agree to a capital injection.

Yes, Japan's attitude to the banks was quite punitive, even as late as the Takenaka plan of 2003. What do you think of that plan?

The Takenaka plan was indeed a most grotesque mistake. Recall then-Finance Minister Takenaka said the Japanese banks' tier-1 capital was rubbish, that essentially too much of it was composed of tax credits. At the time, five years' worth of tax credit could be used as tier-1 capital, whereas in the US, it's limited to one year. Takenaka attacked the Japanese system simply because it was different to the US system.

In fact, the five-year tax credit was the result of painful negotiations between the tax office and the Ministry of Finance. The MOF wanted the banks to write off their non-performing loans with pre-tax income, but the tax office insisted on writing off NPLs with post-tax income unless the borrower has literally gone bankrupt. Eventually, a complicated compromise was reached. The MOF agreed that banks will dispose NPLs with post-tax income, but will be allowed to claim over-paid taxes as deferred tax credit on tier-1 capital. It would become "real" capital once the borrower has gone completely bankrupt and over-paid taxes are returned to the bank. So the banks were incentivised to write off their NPLs—a good thing—which increased the amount of tax credits in their tier-1 capital.

Do you think the US economy will also need huge infrastructure projects, of the type which were heavily criticised in Japan for wasting money?

It will be useful to have huge public works projects to prevent a sharp contraction in GDP. It's the most effective technique, much more so than tax cuts. In a crisis, people save money.

What's your take on the severity of today's banking crisis?

I think it's going to be very serious. I was really shocked when there were no takers for Northern Rock (the British mortgage lender which was taken over by the UK government over Christmas 2007). This is the same as in 1998 in Japan. When Long Term Credit Bank (LTCB) collapsed, it was impossible to find a buyer, either domestic or foreign. That reflects the nature of a systemic banking crisis - when everyone is affected at the same time, there will be no takers, no buyers for stricken banks or their assets. Nationalisation is thus the only way to go, or the whole system will collapse. The fact that Northern Rock ended up in that situation shows how battered the banking sector is. It's not just the US, it's an Atlantic system problem, so it could be far worse than in Japan.

Can you give a figure for the scale of the problem and the extent to which it has affected Japan?

Fortunately, Japan seems to have dodged the worst of the problem. As somebody recently said, "We were not invited to the party so we don't have a hangover." But the costs for the US will not be small. Remember the Savings and Loans collapse in 1989. That cost $160 billion, and it was tiny compared to today's problem. They were small institutions. To recapitalise the banks after the Great Depression, the US injected capital equivalent to one third of the total capital in the banking system. It could well be that much again this time.

What's the nature of nationalising a bank?

Under nationalisation, shareholders are wiped out, so it's not popular with shareholders. Northern Rock (the British bank that went bust), however, already has negative equity anyway. Then, the government takes over, and actually runs the bank. That's vital for the security of the whole financial system, because the bank has millions of arrangements with hundreds of counterparties around the world relating to its business. That's actually the most important part of nationalisation—reassuring counter parties that deals will be honoured. When Long Term Credit Bank (today's Shinsei Bank) was nationalised, the Bank of Japan sent out scores of its staff to reassure counterparties across the world that the derivatives portfolio of LTCB would be honored. That was one of the biggest worries—what might happen if a huge player in the market collapsed. In the savings and loans crisis, the Resolution Trust Corporation was not interested in saving the institutions because they were all so small. Its mission was to sell off the assets of the lenders.

Is the role of real estate in the US similar to the role of real estate in the Japanese bubble?

Yes, the Tokyo area housing bubble in the late 1980s is exactly the same magnitude as the US housing bubble today, but the commercial real estate problem in Japan was far bigger than the one facing the US today. In both countries, prices for real estate went up far too high. As a result, we are in for a pretty bad couple of years. If you look at the US housing inventory ratio, it is at an all time high. For that to return to normal will take at least two years.

Do you feel there is little point in trying to rescue the housing sector?

Yes, forget the housing market. Prices must return to sustainable levels. The priority now is to make sure GDP doesn't fall. The US government must not follow the fiscal example set by the Japanese government, which was dreadfully stop-go. As soon as the economy saw an improvement, the government would cut the fiscal stimulus. I believe the US should have at least a five-year seamless programme of fiscal stimulus. After five years, when the private sector is in better shape, the government can take its turn at fixing its balance sheet.

In retrospect, how did we get into the crisis?

It's because US companies stopped borrowing money after the IT bubble burst. What we are seeing now is the second stage of the bubble. Alan Greenspan (former chairman of the Federal Reserve) had a plan to help the economy out of the IT bubble, namely to stimulate a housing bubble through low interest rates, during which time companies would fix their balance sheet, and eventually resume borrowing, allowing rates to increase and the housing bubble to gradually shrink. But what happened in the US was that companies did NOT borrow money even after their balance sheets were repaired. Managers were too traumatised. On Wall Street, there was consternation. The cost of funding was going up, 'normal' people had bought their houses and companies were not looking for financing. So Wall Street found a new market, the subprime market, $1 trillion went into the subprime market in two years, because the borrowers were willing to pay high interest rates on their homes.

In the recession of the early 1990s, Greenspan saved the banks by giving them a "fat spread". This means keeping funding rates at rock bottom while keeping bank lending rates up. If you give 300 basis points per year to the banks, and do it for three years, you have 9% capital right there. However, that's not possible today, because there is not enough demand to play the fat spread game in the US. Companies and consumers are already too indebted to be willing to take on more debt. That is why capital injection is necessary.

Silicon Valley State of Mind

(enter your email)
(enter up to 5 email addresses, separated by commas)

Max 250 characters

blog comments powered by Disqus