Carry Trade Primer

Well, first off there is Howard Simons, and this little presentation here. (Here's a better link maybe). Charts to come.

‘JPY Carry Trades’ the Undeserved Scapegoat
March 09, 2007

By Stephen Jen | London

Summary and conclusions

Reading the news reports last week, one could be excused for believing that the fate of civilization depended on the so-called ‘JPY carry trades’. In my view, the role that these trades have played in the global risk-reduction in the past week-and-a-half has been grossly exaggerated. These trades in JPY matter for the JPY, but should not matter for other asset markets.

In the near term, I believe USD/JPY and EUR/JPY are likely to rise rather than fall, as new JPY shorts are re-established. Over the medium term, I continue to look for (i) the BoJ to be more hawkish and more assertive and (ii) financial market volatility to rise. Until these two pre-conditions are met, our medium-term forecast for USD/JPY to trade down to 112 by end-2007 and 108 by end-2008 may be in jeopardy. In other words, while I have been structurally bullish on the JPY, ironically I have been unconvinced that the latest correction in USD/JPY and EUR/JPY will be sustained. In the short term, the risks to the JPY are biased to the weak side, in my view.

Not clear why equity investors are so focused on the JPY

I have been surprised by the focus on the ‘unwind of the JPY carry trades’ by many investors and commentators, particularly those outside the currency space. According to press reports, these ‘JPY carry trades’ are highly leveraged, allowing low interest rates in Japan to support global financial prices, regardless of what the Fed and the ECB’s policies are. This view of the world, that massive JPY shorts are financing long-Turkey and long-global equities positions, also implies that with any shock to this delicate situation (e.g., BoJ rate hikes or a rise in uncertainty), these JPY shorts would be violently unwound, triggering a collapse in global financial prices.

I have been puzzled by this view of the world, and believe there are some misconceptions about the ‘JPY carry trades’:

• Misconception 1. The unwinding of the JPY carry trades played a key role in the recent equity market sell-off. In the first five days of the recent risk-reduction, some US$3.0 trillion in equity market capitalization was wiped out. One estimate of the total stock of the JPY carry trades, suggested by Mr Watanabe, Vice Minister for International Affairs of Japan’s Ministry of Finance (MoF), is around US$150 billion. The total equity market capitalization for the world is around US$43 trillion. In other words, the size of the ‘JPY carry trades’, estimated by Mr Watanabe, is equivalent to around 0.36% of the world’s equity markets. The decline in the world’s market cap was already 20 times the size of the total stock of this type of JPY carry trade.

Further, even though the AUD, NZD and INR are arguably the most popular longs for Japanese retail investors, these three currencies weakened last week, but did not underperform other currencies as much as they should have if the unwinding of JPY carry trades were the key driver of global instability.

Blaming the unwinding of the ‘JPY carry trades’ for causing havoc last week is almost as bad as blaming the sell-off on the Chinese equity markets. To me, what happened in the last two weeks was the result of excess leverage, extreme mis-pricing in some markets and investors having the same trades. JPY shorts may have been a popular trade, but it is by no means a dominant position in the market, in my view.

• Misconception 2. Japanese investors hold leveraged interest rate-sensitive foreign currency instruments in massive amounts. This is not correct. First, the figure US$150 billion of ‘JPY carry trades’ I mentioned above is a measure of the total stock of foreign mutual fund holdings by Japanese retail investors in the form of bonds. This is big but not huge. Even if there was some leveraging to financing these positions (though there are no hard data proving this point), the size of the leverage couldn’t be that big. Thus, these Japanese outflows are more ‘non-JPY longs’ than ‘JPY shorts’, and do not really deserve the term ‘carry trades’.

Second, the returns on many overseas equity investments have grossly exceeded the 4.75% of the cash yield differentials. The motivation for Japanese retail investors to go overseas is as much structural as it is cyclical, and as much capital gains-driven as it is interest rate-driven. Using the term ‘JPY carry trades’ to refer to these equity plays is inappropriate, in my opinion.

Third, there are foreign investors, especially hedge funds, who have more recently started to accumulate speculative JPY shorts in size. These are genuine JPY carry trades, i.e., they are interest rate-driven, and are financed by actual leveraged borrowing in JPY. The IMM positioning is a rough indicator of the large size of these positions. While IMM may not show record high short-JPY positions, short-JPY contracts in the past few weeks were nevertheless extremely high (around +115K to +173K contracts) compared with the average over the past decade (around +30K). A violent unwind of these positions could explain the JPY rally in the last week-and-a-half. All this time, Japanese retail investors have been remarkably calm. With the exception of a few hours on Monday (March 5) morning, Japanese institutional investors were net buyers of USD/JPY, not sellers.

In short, the fact that the JPY rallied hard as the risky assets sold off, to me, was a reflection of general risk-reduction, especially by hedge funds. The direction of causality ran from global equities to the currency markets, not the other way around. Further, it was the foreign institutional funds that have been doing most of the buying of JPY in the last two weeks, not Japanese funds.

• Misconception 3. The term ‘JPY carry trade’ is appropriate. It is a catchy term, but quite a misleading one. I have already mentioned above some of the reasons for my objection to this term. However, I don’t dispute that there are meaningful positions in the market that are short-JPY long-everything else, mainly because of the low yields in Japan. The existence of these positions make the ‘JPY carry trades’ a technically correct term, but I believe the use of the term is too broad and sloppy.

• Misconception 4. There is incontrovertible proof of the presence of large ‘JPY carry trades’. On the contrary, data on these ‘JPY carry trades’ are ambiguous. We have written several pieces in the past highlighting the inconsistency in Japan’s data proving that there are large JPY outflows of this type. While some data that track the subscription rates by Japanese retail investors for foreign currency-denominated investment trust funds and uridashi issuances do suggest an increase in outflows starting in late 2005, official data from the MoF and the BoJ on cross-border portfolio flows don’t corroborate this at all (see Further Thoughts on the JPY Carry Trades, February 8, 2007). In fact, the latter data, which should be more comprehensive as they include not only mutual fund flows but also how lifers, pension funds, banks and other institutional funds may be doing, show that, in 2006, there were net bond inflows.

Similarly, there is somehow a popular presumption that non-Japanese citizens (Eastern Europeans in particular) have been taking out their mortgages in JPY. We have looked at this issue more closely, and found that JPY loans outside Japan are small and have been declining, not rising (see Type 3 JPY Carry Trades: Fact or Fiction? November 2, 2006). CHF loans outside Switzerland have indeed become much more popular, but not JPY loans. However, I am puzzled to hear other commentators and investors mentioning this type of JPY carry trade, with no data to prove their claim.

A currency world driven by nominal factors

The fact that nominal factors (cash yield differentials) continue to be more powerful drivers than real variables in the currency space has surprised me. What worries me is that all the reasons I have thought of to explain the dominance of nominal variables — including (1) increased focus on policy rates due to inflation-targeting, (2) enhanced transparency and predictability of central banks, (3) structural capital outflows from aging populations, such as that of Japan, and (4) the greater importance of currency hedging due to financial globalization — are structural in nature.

Risk-taking will recover; an opportunity to ‘upgrade’

With the very positive global economic outlook and abundant global liquidity, I see risk-taking recovering rapidly from the latest correction, though there should be better differentiation between assets than in the past. In other words, this is a great opportunity for investors to ‘upgrade’ their portfolios. Absent other changes, USD/JPY is more likely to reclaim 120 than to trade down to the low-teens, in my view. I am still waiting for the BoJ to become more assertive and hawkish and for the volatility in the financial markets to increase further. These two conditions will need to be satisfied for me to be confident that the JPY will appreciate and stay strong.

Bottom line

I believe the role that the ‘JPY carry trades’ played in the recent risk-reduction has been grossly exaggerated. While there are indeed JPY shorts due mainly to the low yields in Japan, market positioning is not big enough to threaten the global financial markets. The JPY rallying was a result of, rather than the reason behind, the volatility in the equity markets. Going forward, I see much of the speculative JPY shorts being rebuilt, pushing USD/JPY back above 120.

Th FT in March

Carry trades lift hedge fund returns

By Peter Garnham and Gillian Tett

Published: March 14 2007 00:40 | Last updated: March 14 2007 00:40

Hedge fund returns have in part been driven by pay-offs from carry trades in recent years, according to the Bank for International Settlements.

According to BIS research, to be published on Wednesday in its quarterly review, regression analysis suggests that for many hedge fund families, including funds of funds and event-driven funds, there is a significant link between their returns and the expected returns from carry trade strategies.

The report, the first in which the BIS has tried to measure effects of the carry trade on the foreign exchange market, is likely to be widely watched, given a recent focus on the issue. The yen has risen more than 3 per cent against the dollar since the increase in market volatility that began on February 27.

Many analysts say the recent rise in risk aversion caused an appreciation in the Japanese currency as investors unwound carry trades, in which the purchase of the riskier high-yielding assets is funded by selling low-yielding currencies such as the yen.

Before the increased volatility, the yen posted a series of multi-year lows against the dollar, euro and sterling, while the Swiss franc also suffered.

Many analysts point to a massive build-up of carry trades as a prime driver of the weakness in the yen and Swiss franc. However, the BIS says evidence for a recent pick-up in carry trade in its own international banking statistics is mixed. Carry trades can show up in the figures since outright borrowing or lending in different currencies are often hedged in the cash market.

The BIS says that while the rise in the stock of outstanding yen-denominated claims in the second half of 2005 did in part reflect greater lending to UK residents and offshore centres, claims have since fallen.

But evidence from exchange-traded foreign exchange markets shows that speculative bets on yen weakness in futures traded in the US rose between mid-2006 and late February 2007, particularly during periods of yen depreciation.

Merrill Lynch said on Tuesday that its research in the Tokyo interbank market showed that non-Japanese banks’ funding had surged from Y1,500bn to Y7,500bn since the end of 2005.

“This Y6tn ($51bn) surge in non-Japanese funding is a new development that, in our view, points to the true carry of yen-funded leveraged position build-up in the recent past,” said Jesper Koll, economist at Merrill Lynch, who believes “the absolute size [of this carry trade is] maybe around $1tn, with about a third of the positions built up in the past 12 months”.