Please, click here to read this article in pdf format: august-6-2009

I am a bit more confused today than I was yesterday. I was hoping to see a bear flattening in the yield curve, indicating that implementation of an exit strategy by the Fed was in the works. A bear flattener is a move in rates where short-term rates increase faster than long-term rates, with the gap between the two compressing. Seen from another angle, in a bear flattener, short-term bond prices fall faster than long-term bond prices. And this has not happened this week so far. The chart below shows that in fact, the opposite took place and rather violently, in the last trading hours yesterday. What happened? One of the possible explanations is that with today below consensus macro data releases (Employment and ISM Non-Manuf. Composite), the horizon for any sooner rather than later rate increase dissipated into oblivion. Adding fuel to the sell-off, there was also mortgage-related selling:

Chart 1: US Treasury Actives: Yield curve change since last Friday (source: Bloomberg)

Another interesting development in the last days has occurred in the correlation space. We dealt with this a few days ago (www.sibileau.com/martin/2009/07/29 ). We said that investors would be wrong to expect that correlation (of defaults) fall with systemic risk:”…Perhaps what misleads investors here is that systemic risk may not be falling, but evolving, switching from liquidity risk in its purest form (lack of financing) to inflation risk (lack of profitability, as costs rise or don’t fall as quick as revenue, while short-term financing becomes more expensive due to a crowding-out effect created by fiscal deficits)…“. In particular, we took the structural approach and said that: “… If you believe that correlation of defaults is a product of correlation in equities (in their expected returns) and their respective volatility, it will be easier to get the picture…”. Interestingly enough, Bank of America’s Credit Strategy team put a note yesterday on this issue (Situation Room: “Credit 25, VIX 0”), suggesting that while individual stock volatilities (within S&P500) did decline, the VIX did not react because an offsetting increase in implied correlation (in returns within the S&P 500 stock portfolio).
This takes me back to a point I am starting to emphasize more and more lately: Global coordination of monetary policies. Yesterday, we referred to the evolution (decreasing) in the Libor-OIS spread. On this subject, a friend and reader kindly sent us a chart (reproduced below) published by the Bank of Canada in its July 2009 Monetary Policy Report (source: Bloomberg). This chart shows the spread between the main (i.e. US, UK, EUR, CAD) 3-month interbank offered rates and their respective overnight index swaps. At first sight, the correlation among them is impressive. Notice too that the absolute differences among these spreads have been steadily decreasing, suggesting a radical change in the factors driving FX dynamics going forward (=if illiquidity drove the value of the USD up in 2008, this is no longer the case in 2009).