US and Japanese yield curves: What they add up to

With economists divided over whether the United States economy is headed for a soft or hard landing, the debate over the reliability of forecasting models is taking on increasing importance, and probably none more than the debate over the reliability of the yield curve as a forecasting tool. And in a globalised economy, there is little reason to restrict the debate to just the yield curve in the US.

On 24 November, John Fernald and Bharat Trehan, vice president and research advisor respectively at the Federal Reserve Bank of San Francisco, published an economic letter that asked: "Is a Recession Imminent?"

"The yield curve is perhaps the best known of all the indicator models used to predict recessions," they wrote. "Based on data for November 8, the model estimates a 47% probability of recession over the next four quarters."

However, they also wrote that the probability of a recession over the next year may not as high as the yield curve suggests. "There is reason to be skeptical about the current high estimate of the probability of recession, because the unusually low rates at the long end of the yield curve are not well understood," they wrote.

They cited another model developed by Michael Dueker that also included real GDP growth and CPI inflation as variables in a vector autoregression model. According to Fernald and Trehan, this model predicts some further deterioration in business conditions over the next year but "does not see much more than a 10% chance of a recessionary quarter over this period".

Fernald and Trehan also found that surveys of forecasters had estimates of the probability of recession that "are all lower than the one based on the term spread and the yield curve", while financial markets "exhibit little evidence of distress", with the Dow Jones Industrial Average hitting record highs recently and various risk spreads at low levels.

Therefore, they concluded that "while the probability of recession might have gone up somewhat in recent months, it is not yet at worrisome levels", and suggested that "it may be useful to supplement data from the surveys with data from indicator models that attempt to measure the current state of the economy".

Actually, there is a simple way to reconcile the forecast made from the yield curve with the other forecasts -- and without even straying away from the yield curve. Just factor in the Japanese yield curve.

Because Japan is fighting deflation, the Bank of Japan has kept monetary policy easy and interest rates low. This has allowed investors to fund investments elsewhere by borrowing at the lower rates in Japan -- the so-called yen carry trade. This at least partially explains the low interest rates in the United States and elsewhere.

In other words, the low rates at the long end of the yield curve in the US could be a sign of an excessive supply of money, not just of inadequate demand for it. And excessive money supply is normally a stimulant for the economy. Fernald and Trehan have good reason to be skeptical about the high probability of recession based on the US yield curve alone.

So can we still use the yield curve to forecast the US economy? Yes, but we should also take the Japanese yield curve into account. After all, if Japanese monetary expansion tends to invert the US yield curve, US monetary restriction tends to steepen the Japanese yield curve. Adding the Japanese yield curve to the US yield curve offsets some of the misleading bearish signal from the latter alone.

The chart below shows the real US GDP year-on-year growth from 1986 to 2006, and compares it with the spread between the yield on the 10-year government note and the central bank's benchmark rate for both the US and Japan. It also shows how a simple average of the two spreads vary over the period.

The chart shows that the combined spread is actually still in positive territory. So the US economy is possibly in better shape than the US yield curve alone implies.

Two caveats, though. Firstly, the current value of the combined spread is about the same as it was in 2000, just before the 2001 recession; that is not exactly a comforting thought. Secondly, both the US and Japanese yield spreads are pointing south with no sign of an interest rate cut whatsoever from either the Federal Reserve nor the Bank of Japan, which means that the prognosis for the economy could be getting worse.

Finally, if Japan's yield curve moderates the pessimistic message from the US yield curve, Japan's stock market can also arguably be used to dampen the optimistic message from the US stock market. While the Dow Jones Industrial Average is up 14.6 percent year-to-date as at the close on 24 November and is near an all-time high, suggesting strong economic growth, the Nikkei 225 is down 2.4 percent year-to-date based on the same day's close.

That is perhaps yet another reason to watch the Japanese markets.