Via Horseman Capital's Russell Clark,
“The upshot of this is that gold looks better than treasuries. However, it should be pointed out that this does not mean gold is a buy. It only states that gold could do better than treasuries in these market conditions.”
In December 2012, I wrote a market view outlining a potential bear case for gold (Gold – A Thought Piece). This note was making the point that gold follows Asian currency strength, in particular the Indian rupee. And in January 2014, I made the case for buying long dated treasuries (US Treasuries – The Buy Case).
Part of the argument here was that Asian currencies’ weakness tends to imply slower global inflation, making bonds attractive. Asian currencies are a diverse mix, so I tended to use the JP Morgan Asia Dollar Index (“ADXY”) as a proxy. One simple way I think about this relationship is to compare ADXY to gold/TLT ratio. (TLT is a long-dated Treasury ETF). As can be seen, the relationship tends to be quite solid, except recently when gold has started to perform better than TLT despite the ADXY being still lower than a year ago.
The logic behind this line of thinking seems quite sound to me, so what explains this recent divergence. I have the currency weightings used to calculate ADXY as of 31 December 2012. At this time weightings to CNY and HKD made up 47% of the index. Both the CNY and HKD have managed exchange rates, CNY on a crawling peg while HKD is a fixed peg to the dollar. The remaining currencies tend to be free floating, and should react much more readily to changing investment conditions. I have created an adjusted ADXY index, where I have removed CNY and HKD from the index. The adjusted index seems to suggest that gold/TLT ratio relationship to Asian currencies still holds true.
Why does the CNY deviate from other Asian currencies so much? As pointed out in a recent market view, the CNY began to weaken after many other Asian currencies. This suggests that CNY will strengthen after other Asian currencies.
The upshot of this is that gold looks better than treasuries. However, it should be pointed out that this does not mean gold is a buy. It only states that gold could do better than treasuries in these market conditions. Gold could still fall from here, but treasuries could fall further. One way that this could occur is a collapse in the US dollar, and the Federal Reserve reacting to a weakening US dollar by raising rates. Certainly, the US dollar is very expensive versus historical levels.
The other possibility is a bond bear market as foreigners dump treasuries. The US net international investment position has deteriorated significantly over the last decade.
A large part of this has been driven by an increase in foreign holdings of treasuries.
So, what does it all mean? It is very suggestive that the US dollar has peaked, and hence gold in US dollar terms should do well. It also suggests that long dated bonds could be much weaker than expected. This creates a conundrum, as higher yields tend to be dollar bullish and bad for gold. My view is that US yields have been kept suppressed by foreign investors. European and Japanese investors have been forced into treasuries by the negative interest rate policies of their central banks, while Chinese investors have reacted to devaluation fears on the CNY by also buying US treasuries.
The Trump administration has sought to reduce US bilateral trade deficits, and to my mind is suggestive that US policy may seek to weaken the dollar. Given the large negative net international investment position, and already falling holdings of treasuries, perhaps foreigners selling US Treasuries would have the net effect of forcing both the US dollar and bonds lower? I would be very cautious with investments that rely on either a strong US dollar or low bond yields.