The chart below shows the UST yield curve over the past 20 years: as is more than obvious, every single point left of the 10 Year is at record tights. The only question on everyone's lips is where do we go from here. And that is where the confusion really hits. The confusion is further intensified by the sudden collapse in the 2s10s and the 2s10s30s butterfly. The odd thing here is that a flattening move as violent as recently seen in these two curves, has historically preceded a rise in the Federal Funds rate as can be seen in the chart to the right, before the Fed began tightening in 1999 and in 2004. In other words a flattening has traditionally been a leading indicator to an economic improvement (as liquidity extraction tends to go side by side with a pick up in inflation and thus economic growth). Alas, this time around, a tight monetary policy is the last thing on the Fed's mind, and the economy is only starting to demonstrate it is rolling over into a second and more violent recessionary round. In essence, the Fed's interventionist intention of purchasing the entire curve (including the long-end), as recently announced by the FRBNY, has completely dislocated all leading signaling by the curve itself. As a result, speculation is now rampant as to what may or may not happen. A case in point are the divergent opinions of Bank of America and Morgan Stanley. While the former Merrill Lynch is advocating an outright 10s30s flattener, Morgan Stanley is sticking to its guns and continues to push for a steeper curve: this in spite of the collapse in the 2s10s from a records steepeness of almost 290 bps in May, to under 220 bps as of Friday's close: the over 25% collapse is enough to blow up most of the funds who had positioned themselves for further steepness. At least Morgan Stanley is consistent. Yet both banks urge clients to hedge their trades and provide creative ways to do so, as both realize the likelihood of being wrong, now that the Fed is openly the biggest market participant, is probably higher than the inverse.