Will a US recession strengthen the dollar?

Macro Man is in a bad mood this morning. Having enjoyed a relatively hot hand for most of the second half of last year, he's started 2008 ice-cold. Every trade he's done so far this year has been a loser- long silver, the US/Euro swap spread, and the FX beta plus carry basket. The latter was exited early this morning....and now it is coming back hard. The upshot is that Macro Man frittered away a strong start to the month and is now in the hole. While ideas are percolating in his head, his cold hand dictates a degree of caution in throwing risk at the market.

The performance of the dollar since 8.31 EST on Friday has certainly been interesting. The buck had every reason to get caned, and while it certainly gapped that way it's subsequently come back to retrace all of its post-payroll losses and more against the G3. What's going on here?

One possible explanation is an emerging school of thought that a US recession/quasi-recession is actually good for the dollar. According to the proponents of this theory, weak/negative US growth is both damaging to the rest of the world and a catalyst to encourage US investors to bring money back home. The upshot is that there is less demand for foreign assets/currencies and more demand for US assets/currency; hence, the dollar rallies.

Morgan Stanley is among the adherants of this view, with Stephen Jen using a framework known as the "dollar smile." According to this construct, in the event of very weak US growth the dollar goes up for reasons cited above. In the event of very strong US growth, the dollar goes up because of the attractive return on US financial assets. Only in the middle, trend-ish growth scenario does the dollar fall, as investors have sufficient appetite to take risk but US returns are unappealing compared to the rest of the world.

It seems remarkable to consider that a country with external deficits the size of the US could see its currency rally in two of the three "states" described above. Of course, the trend-ish growth scenario can well comprise the lion's share of the time spent in aggregate. Nevertheless, it is probably worth checking to see if the notion that a US recession is bullish dollars holds water.

Macro Man constructed a simple study wherein he looked at the 4 quarter moving average of the normalized deviation of quarterly GDP growth from its ten year trend. He then compared this reading with the subsequent one year change in the Fed's broad USD TWI. The results, while not conclusive, nevertheless represent something of a victory for what one might call common sense. Over the last 25 years, the correlation between GDP growth and the subsequent change in the dollar is 0.4. In other words, when the US economy slows, the dollar tends to fall afterwards. Call it the "dollar sneer."
Of course, the degree of lag appears to vary, which is probably down to the speediness of the monetary response of the Federal Reserve, among other things. Taken at face value, the strong US growth from mid-2007 would suggest that the near-term outlook for the dollar is actually fairly neutral. Yet a GDP forecast profile that's probably reasonably close to consensus suggests that we should expect pressure on the dollar to resume. The fact that real Fed funds rates are already zero offers a powerful disincentive to hold dollars; by the end of the month, real Fed funds will likely be negative.

None of this is to say that a period of US growth of 1% or lower means that the dollar will fall against all things. Weak growth, particularly if it spreads, should have an impact. Macro Man's rule of thumb is that when the poo hits the fan, the money comes home. This is the "left side" of the dollar smile described above.

And who, pray tell, has been the primary recipient of "hot money" capital inflows over the past year or two? Emerging markets and Europe. It's not terribly difficult to construct a scenario wherein weak US growth broadens, and the currency game finally abandons growth (or more to the point, its first cousin, carry) and focuses on value. If so, then the dollar is not the obvious trade, given that it is undervalued against some currencies and overvalued against others. Surely the real trade is to sell those currencies that are overvalued against everything (EUR and GBP) and buy those that are undervalued (BWII currencies and yen)?

Last week, Macro Man highlighted the break of key suport in GBP/JPY. That break has held, though if the MPC doesn't cut rates on Thursday there may be another squeeze on the cards. Another egregiously mispriced pair, and one where there is a political will to see it move closer towards fair value, is EUR/RMB. It remains above its uptrend line since the end of 2005, but that support isn't massively far away. A move to the middle of its range since the euro's inception (hardly something that is out of the question from a fundamental perspective) would yield a move of 15.65%, less carry.Macro Man is somewhat positioned for a "value matters" regime in currencies, via his EUR/USD powerball strip and the short USD/JPY cash position. There is ample scope to add. However, given his cold start to the year and the potential for a further squeeze this week, he'll wait for more attractive levels before layering fresh risk.