The yen bears have been frustrated by a series of developments. They were unpleasantly surprised by news that the Japanese themselves were large sellers, not buyers, of foreign bonds in the first full week of the new fiscal year. In addition, more often than not, since the BOJ’s announcement, the yen has strengthened, not weakened, in the Tokyo trading session. And the yen bears were unable to absorb the yen buying that capped the US dollar just below the psychologically important, JPY100 level.
Another factor that went against the yen bears was the backing up of long-term Japanese interest rates, even though the BOJ is planning on buying some 70% of amount of JGBs that the government will issue until at least the end of next year, seems counter-intuitive to many. Yet, in the most fundamental sense, the Q-squared policy (quantitative and qualitative easing) works through raising inflation expectations.
Nominal interest rates can be understood as a combination or real rates and a risk premium. Since Japan has a yen printing press and a fiat currency (where the supply is not limited to gold or silver holdings), the risk premium is not that of a default. In fact, the cost of 5-year insurance, through the credit-default swap market, has actually fallen about 5 bp to 70 since the BOJ’s announcement.
If the BOJ’s policy is going to work, one would expect to see higher interest rates as the market demands compensation for the increased risk of inflation. The risk premium is to cover the possibility that the BOJ succeeds in igniting inflation.
Nevertheless, some observers seem to think that because the BOJ is buying bonds, it does not want yields to rise. To the contrary, as long as the move is orderly (which, thus far, it does not seem to be), an increase in interest rates is not undesirable. This seems especially true if the stock market can continue to rise and the yen remains soft.
This appears somewhat contradictory. Over the past sixty days the yen (not the dollar) and the Nikkei moved in opposite directions in roughly 54 sessions. And this has been fairly steady on a 60-day rolling basis since the start of the year.
The backing up of Japanese bond yields, and the decline in US yields (partly in response to disappointing US economic data) has narrowed the interest rate differential. The 10-year spread, which we find tracks the dollar-yen exchange rate, has narrowed considerably, and this is yen supportive.
When the Japanese election was called in mid-November 2012, the US was offering about 85 bp more than Japan on 10-year notes. The spread widened to about 140 bp by the middle of March, as the dollar rallied. It has been drifting lower since, but spread compression accelerated in the past week.
Before the BOJ’s announcement, the spread was near 130 bp. It will start the new week near 110 bp, the narrowest since late January. This should not be understood to mean that the dollar is going back to where it was then (~JPY90). We do not find a one-to-one correspondence between the interest rate spread and a specific dollar level. It is the change and direction that we find most important.
The dollar closed the week poorly against the yen. It was the second consecutive lower close in the North American session. The dollar has only fallen three times this year for three consecutive sessions. While the market may feel set for a correction, the technical indicators, like MACDs, the and Bollinger Bands do not show an overly stretched market, though the RSIs (14-day) have turned lower.
The most important indicator next week may be how the dollar performs in Tokyo at the start of the week. If investors there do not sell into the pre-weekend yen bounce, then the offshore yen bears will have nothing to hang their proverbial hats upon. If the yen was previously sold (and bought other assets) in anticipation of what Japanese investors would do, and the Japanese do not respond as expected, the shorts are psychologically more vulnerable and in weak hands. The move back above JPY99.00 before the European market open may serve to steady the nerves of the yen bears. On the other hand, the next retracement level comes in near JPY97.25.
The euro’s upside correction remains intact. A convincing move above the $1.3115-20 area will signal the start of the next leg up that is worth at least another cent and maybe 2. Ideally, the euro holds above the $1.3020 area, but it may require a break of $1.30 to undermine the correction.
Sterling had a poor close before the weekend, but inside Thursday’s range. Sterling’s advance has stalled just in front of the resistance we identified near $1.5425. If the bullish flag pattern is to remain intact, sterling should not see much in follow through losses. If these materialize, sterling may have to slip back toward $1.52 to test support.
Just like we are initially reluctant to read too much into sterling’s poor performance on Friday, so too with the Canadian dollar. It had, after all, reached it best level in nearly two months on Thursday, before posting an inside day and a close near session lows. However, we do recognize that the US dollar bottomed at a technical retracement objective near CAD1.0080. If the CAD1.0180 area does not hold back upticks, it would suggest the greenback’s downside correction is over. Over the medium term, we look for CAD1.05.
The Australian dollar reached almost $1.06 before the unwinding of cross positions against the yen took a toll (as it appears to have done on the other major currencies as well to varying degrees). Some observers linked the heavier tone of the Aussie before the weekend to the slide in steep slide in gold prices.
While possible, our work shows a very weak correlation between Australian dollar and gold. In fact, on a rolling-60 day basis, using percentage change, the correlation is at its lowest since Q4 ’11 just below 0.20. Running the correlation on the level of the Australian dollar and the level of gold, correlation is roughly the same.
The Aussie should find support in the $1.0440-60 band and this will allow it to challenge its 8-month cap near $1.06. Similarly a break would warn that the cap once again held and the Australian dollar may be headed back into the $1.0300-50 area.
While we continue to like Mexico’s fundamental story, the technicals are beginning to look stretched and the dollar has approached a key psychological level of MXN12.00. Given the popularity of a short yen and long peso position, either separately or together, if one expects the correction in the yen to continue, than a short peso position might be an interesting way and a less volatile way to express that view.
Lastly, the most striking observation about the speculative positioning in the futures market is the reduction of exposure. Shorts covered and longs were liquidated. The two minor exceptions was the inconsequential increase in gross short sterling positions and the small increase in gross Canadian dollar shorts. In chess, it is said, when in doubt push a pawn. In the trading, when in doubt, move to the sidelines.
week ending April 9 | Commitment of Traders | |||||
(spec position in 000’s of contracts) | ||||||
Net | Prior | Gross Long | Change | Gross Short | Change | |
Euro | -50.9 | -65.7 | 35.3 | -9.1 | 86.2 | -24.0 |
Yen | -77.7 | -78.2 | 39.7 | -7.3 | 117.4 | -7.7 |
Sterling | -70.0 | -65.0 | 27.1 | -4.7 | 97.1 | 0.2 |
Swiss Franc | -10.0 | -12.0 | 6.6 | -3.2 | 16.6 | -5.2 |
C$ | -71.1 | -64.5 | 24.1 | -2.2 | 95.2 | 4.4 |
A$ | 77.9 | 84.0 | 117.5 | -18.0 | 39.7 | -11.9 |
Mexican Peso | 143.0 | 143.0 | 151.8 | -10.4 | 9.2 | -10.2 |