Goldman Explains How It Can Be Long The Dollar AND Call For 2+ Years Of No Fed Fund Tightening
The biggest discrepancy coming out of Goldman these days is not the "conviction buy" external rating on all equities, even as the prop guys keep selling, but the conflicting opinion of a strong dollar coupled with expectations for two+ years of no Fed Fund rate increases. Conventional wisdom of course says that it has to be one or the other - can't have both. Apparently Goldman clients have voiced enough shock and awe with this position by the masters of the universe, that it has provoked the firm to come out with a clarification of how its trading desk could have been axed so wrong. Below is the proposed justification.
Many clients have asked how we can justify our forecast of Dollar strength on a 12-month horizon (EUR/$ 1.35) given that we now expect the Fed to remain on hold right through 2010 and 2011. The answer to this question is twofold. In terms of rate differentials, strong shorter-term correlations do exist between rate differentials and currencies to the extent that both reflect changes in the growth outlook. However, over longer horizons, these correlations tend to be very instable and often dominated by trends that point in opposite directions.
For example, there is no clear long-term link between the trade-weighted Dollar and the 2yr rate differential between the US and its key trading partners, as can be seen in the chart. In 2001/02 rates moved against the USD, while the USD strengthened. From 2003 to 2005 rates moved in favour of the USD but the Dollar weakened. And, finally, in 2008 the Dollar rallied strongly while rate differentials remained unchanged.
On the other hand, we have shown on several occasions in the past that capital flows do have a pretty robust impact on the USD. In that respect, it is critical to focus on the BBoP, which has not yet improved because the narrowing in the trade deficit has been offset by deteriorating portfolio flows so far. But once the BBoP improves, the Dollar will likely strengthen on a tradeweighted basis and this is not necessarily linked to Fed hikes. In particular, we can imagine a scenario where investors start to re-invest in the US on the basis of Dollar undervaluation, reduced US imbalances and early pockets of sustainable demand growth, when at the same time the Fed continues to worry about a high unemployment rate and keeps policy on hold.
We are not really sure when the BBoP will improve sufficiently to change the USD outlook. It certainly is too early now, and this is why our 3- and 6-month forecasts remain firmly Dollar-bearish across most major currencies. But towards the second half of next year, there could be a chance of improving capital inflows into the US, which is reflected in our 12-month forecasts. We often think of our 12-month forecast as a GSDEER reminder, while we wait for evidence of improvements in the notoriously volatile and difficult-to-predict capital flows.
It is also worth pointing out that correlations between short rates and FX are currently a lot weaker than correlations between equities and FX. Again, an indication that too much focus on rate differentials is not particularly useful when assessing the outlook for the Dollar.
Then again, before you rush out and buy dollars (which is probably not a bad idea but for other reasons), Goldman reminds us just how badly it has fared in its prediction of Yen levels.
We have been wrong with our 2009 $/JPY top trade. When we made the case for a long $/JPY position last July, we were aware of the high sensitivity of the Yen to US rates. Since about 2005, the rule of thumb was that $/JPY changes by about 1% for every 6bp move in US 2yr swap rates (see chart). We had discussed extensively that Japanese hedging activity was probably one of the key drivers of this relationship. We also made the case that $/JPY positioning had come back to more neutral levels after many months in long Dollar territory.
Based on this assessment, the likelihood of a rise in $/JPY appeared to have increased substantially. With the Fed on hold for an extended period, the risks of a move lower in US rates seemed fairly muted given that 2yr USTs only yielded a bit more than 1%. And with the Japanese data so weak, we did not expect markets would significantly increase long Yen positions. Moreover, long Yen risk had been a popular cross asset hedge during the 2008 carry unwind and post-Lehman slump—another reason not to expect a renewed increase in long Yen risk.
But things played out differently. First, we plainly misjudged the impact of the newly elected DPJ government on the JPY. We still think most of the perceived change in FX stance towards welcoming a stronger Yen was linked to teething problems on the communication side, but the new government allowed markets to think that a genuine shift may have happened. In response, speculative positioning shifted substantially into long Yen territory, contrary to our expectations. Second, US rates did decline further, troughing at only slightly more than 60bp for the 2yr US Swap rate after the Dubai news hit the tapes. While the decline in US rates during the second half of the year was small compared with the decline in the previous two years, it was sufficient to push $/JPY substantially lower.