Monday, April 2, 2012

US Treasuries: A great hedge or the Biggest Bubble of All Time?

This is a question that has been repeatedly asked of me and other researchers at RiXtrema. US Treasuries is a 'spot' is where the past observations and future predictions meet and do battle. Why?

The Safe Haven Argument
It is really difficult to come up with an economically sound and clear argument as to why UST must remain a safe haven. The 'least dirty shirt' quote of Bill Gross (not one of the UST fans, to be sure, even to his detriment over the last couple of years) is really not an argument as much as it is a restatement of present realities. Kind of like, 'It is what it is'. He says: “The world is full of dirty shirts in terms of excessive debt, and the United States is one of those countries, but it still remains the reserve currency and still remains the flight to quality haven.” This argument is a bit circular, something like "UST is a save haven, because it is a flight to quality haven"; a premise presupposes the conclusion. There is no denying that historically UST has been a great hedge for instability, its safe haven status (the ultimate safe haven in the dollar driven financial system) clearly reinforced by its performance in crises over the past 4 years. If we had to come up with one solid argument for why UST is a safe haven and has remained one despite all the talk of a bubble, it would be simply that the financial system needs a safe harbor and all the liquidity must go somewhere in crisis. UST provides a big enough container for all that liquidity to flow into. There is also the psychological habit here reinforcing the 'container theory'.

The Biggest Bubble Argument
However, there is something that makes this picture increasingly blurry. The premise of RiXtrema's Allocation Model is that all financial markets go through long periods of risk underpricing (boom) to short periods of violent risk overpricing (bust). Risk underpricing is neither unhealthy nor peculiar to the financial system. There is a ton of research in behavioral and sociological disciplines showing that people overreact to crises, but then forget them fairly quickly (e.g. buy most flood insurance right after the flood when it is statistically least likely to reoccur and buy less over time, as the likelihood goes up). This behavior is actually 'healthy' to some degree, because we wouldn't be able to act otherwise and would be always paralyzed by the memory of prior catastrophes. The problem appears when risk pricing deviates dramatically from historical median levels, thus setting up the stage for the bust. This could be enabled by various financial engineering innovations that bring in the 'new era' and/or flow of cheap/free liquidity (purely hypothetically speaking, of course). The key problem of risk management is that risk can be mispriced for a long long time before a crash ensues. Thus, many investors call the crash too early, despite intuitively (or even quantitatively) capturing the problem in risk pricing. In fact, these stages of dramatic exuberance could be the most profitable stages of the boom cycle.
Has risk been mispriced in the UST in a dramatic fashion? Most certainly, yes. At RiXtrema we use a number of metrics to measure risk mispricing (all percentiles, with 100 being the most dangerous for the investor/risk manager). Some of them are:
Short Term Capital Flows/GDP (short-term capital is just about the most dangerous thing that can happen to a sovereign economy)
Real Effective Exchange Rate
Gross Leverage
Etc. etc. (see The New Paradigm of Risk Management whitepaper for more on risk mispricing and loss of conviction as applied to different markets)
According to those measures, the risk in UST is dramatically mispriced (click the picture below)

As you can see from the table US Bond is lighting up quite a bit (yellow and red indicate above median and the upper quartile). However, that has been happening for a while and yet our model never showed increased risk for UST over the past two years. That is because risk mispricing by itself is a necessary but not a sufficient condition of the crash risk.
On a related note, it is worth noting that we wrote about this almost exactly 2 years ago, on April 2, 2010:
"Though the situation of the US government is dire, it has a major asset in the US dollar, which still remains the world's reserve currency. Other countries do not have such a golden goose. Therefore, as the current slump continues we are likely to see other sovereign debt crises a la Greece (this part of the article was written before the Portugal downgrade) in the shorter term. These sovereign crises will produce a temporary strength in the dollar. This will likely happen before we get to the point of high interest rates in the US and a partial default by inflation."
Needless to say, that is exactly what we have seen. But how long can UST temp fate? The short answer is that until something we call 'loss of conviction' sets in, the UST will not be in crisis. More to follow in the next post on the concept of loss of conviction and why it is so crucial.

No comments: