Japanese government bonds: could this be the best trade ever?

The Holy Grail for investors is a trade with big upside but little risk. Of course, they are very rare. The closest I've come was the dollar in March 2008. At an exchange rate of two to the pound, it was one third undervalued compared to the long-term average. I could not believe it would fall much further.

Triggered by the credit crisis, the dollar duly rose 40% in the next 12 months. And my money market fund paid 4% interest along the way. Over the same period, stockmarkets lost half their value.

Adjusted for risk, I reckon this is my best-ever trade. Until today, that is. Now I think I've found an even better one. Take a look at the 2011 Japanese government budget, below. Tax revenues don't even cover debt service and social security. New borrowing and transfers from savings pay for more than half of total spending. At 637trn, total outstanding debt is more than 15 times tax revenue. Yet the government is paying an interest rate of barely 1.5%. Just imagine what this budget would look like at rates of, say, 6%. You'd have to add another 30trn to spending. Then revenues would barely cover interest payments.

See also

• Japan: classic investment opportunity, or too dangerous to touch?

How on earth can the Japanese government borrow 15 times income at an interest rate of just 1.5%? Its trick is to sell bonds to its citizens. Most Japanese government bonds (JGBs) are owned by banks and pension funds. And it's worked out well for them. The yield on the ten-year JGB fell from 6.75% in November 1989 to below 1% in 2010. That's a capital gain of 53% on top of 20 years of interest income of 6.75%.

But the trick is about to stop working. As the Japanese people age, they need to sell their investments to pay for retirement. In 2009, Japanese pension funds were net sellers of government bonds for the first time in years. And they'll keep selling as more retire. The government is running out of buyers for its debt. That means it will have to offer much better yields. The price of JGBs is going down.

Here's where it gets interesting. You see, yields can't go much lower. The absolute limit is zero. The all-time low was 0.43% in May 2003. And I sense the practical limit is now not much less than 1%. So, if you could take a short position, there's limited downside risk.

Japan's finances are in terrible state. So I reckon it should pay a premium over UK gilts, for example. Our ten-year government bond yields 3.7% today and is probably headed higher. Inflation is picking up all over the world. So I think JGBs will soon yield 5-6%.

If I'm right, a spread-bet to sell the ten-year JGB could make 48 times your initial stake. If you're new to spread-betting, I show you how to place a bet below. And for the technically minded, I explain what you're betting on below.

But first, let's examine the risks and the possible pay-offs. Have a look at the chart below. The solid red line shows how the price of the bet changes with the yield on JGBs. The dashed black line shows your profit or loss per £1 bet.

At the market close last Friday, the price of the bet was 13,923 to sell. That equates to a yield of 1.58%. If I'm right and the yield rises to, say, 6%, the futures price would fall from 13,923 to 10,000. You'd make £3,923 per £1 bet. But it costs you £8 to close out the bet. So your net winnings would be £3,915. Your minimum initial deposit is £81 per £1 bet. So your profit is 48 times.

You'll probably not be quite that lucky. There's no guarantee that the yield will go straight up. If it were to fall in the short-term to, say, 1.0%, the future would rise from 13,923 to 14,565. You'd be down £642 per £1 bet. So you'd need to deposit another £642 on top of the initial £81. That loss is not realised until you actually close out the bet. But if you can't afford to lose this sum, don't bet.

My sense is that 1% is realistically the worst-case scenario. So if I'm right, you'd make £3,915 profit from a maximum investment of £731. That's a reward to risk ratio of overfive to one.

Of course, yields could fall further. The all-time record low was about 0.5% during the worst period of deflation in 2003. But I find it hard to believe yields will retest previous records. Rebuilding the economy after the terrible tragedy of earthquake and tsunami will only make government finances even worse. If you're worried about profiting by shorting Japan, buy Japanese equities as well (as my colleague Merryn Somerset Webb suggests here: Japan -classic investment opportunity, or too dangerous to touch?) Before last Friday, I considered Japanese equities so undervalued that you might double your money over the next three years. These days, you might make slightly less. But now you're a net provider of capital to the country.

I can't tell you when JGB yields will rise. It could start next week. Or it might take several years. But shorting the JGB future via a spread bet offers the potential return of five to one, maybe a lot more. It's worth the wait.

What am I betting on?

You're not actually shorting the JGB itself, but a three-month future contract on the JGB. This might sound more complicated than it really is. A future contract is an agreement to buy or sell a security at a fixed date in the future, at a price which is fixed today.

In this case, you contract to sell the JGB on 8 June at a price of 13,923. No money changes hands until settlement day 8 June, in this case. However, you must place a minimum sum on deposit as a sign that you are good for the money. IG Index requires a deposit of £81 per £1 bet.

If the price of the JGB future goes straight down, £81 is your total stake. However, if the price of the JGB future rises in the short term, your bet will be out of the money. So you'll have to deposit additional sums to cover your unrealised losses.

The future contract is based on a notional ten-year 10,000 bond paying a coupon of 3% every six months. This means that if the yield were 6%, the price would be 10,000. Actually, the current yield is 1.58% and the mid-market price is 13,927.

The futures price differs from the bond price in two ways. If you own the future, you don't receive the interest payments. That makes the future worth a bit less than the bond. Second, you don't pay upfront unlike buying the bond itself. That makes the future worth a bit more than the bond. The precise difference depends on how many days are left until settlement day, and how many days to the payment of the next coupon.

How to place your bet

Spread-betting can be daunting if you have no experience. But it needn't be. I have an account with IG Index so I can tell you how to place the bet with them. Other spread-betting companies are similar but the details may differ. You can also check out the instructional videos on their websites. But if you are uncertain in any way, don't trade. Spread-bets are highly geared. So the profits or losses can be large, relative to your deposit.

After you've opened your account, you'll find the JGB bet in the section marked 'Bonds and Money Market'. Click on 'Asian' and then 'All'. At the market close last Friday, the price was 13,923 to sell and 13,931 to buy. The bet also has a time limit. At the time of writing, it expires on 8 June 2011. At that time, it will be replaced with a new bet expiring three months later.

Since I don't know exactly when JGB yields will rise, you need to choose to have your bet roll over automatically. To do this, go to 'my account' 'settings' 'rollovers' and make your selection. You will suffer rollover costs every three months. With yields at 1.5%, the costs are modest, compared to the potential profit. But your profits will depend on how long the bet takes to pay off.

The best time to place the bet is first thing in the morning, UK time. The market shuts at 9am. But if you are familiar with limit orders, you can also place a bet when the market is closed.Click on sell and the deal ticket will open up. At this point, you must decide the size of the bet you'd like to make. The minimum is £2 per point. If that's what you choose, you'll have to deposit a minimum £162 to open the bet. Then you'll make £2 profit for every point that the price falls below 13,923. But you'll incur a bid-offer spread of £16 when you eventually close out the bet.

Track how you're doing in the window marked 'Open Positions'. And close out the bet at any time you like. Just open up the deal ticket in the 'Open Positions' window and click on 'Close Position'. It sounds a bit complicated, but it's simple enough in practice.

Simon Caufield writes the True Value newsletter.