At the end of last year the 30-year swap spread (the spread between the 30-year interest rate swap fixed rate and they yield on the 30-year Treasury bond) turned negative. It reached levels as low as negative 60 basis points. It was a technical issue more than anything fundamental. Two theories persist as to what caused this anomaly:
- Institutions who owned fixed rate bonds (corporate and agency) had to sell to raise liquidity. Many had these fixed rate securities hedged with swaps. As they sold the securities, they had to unwind the swaps in size, diving down the fixed rate relative to treasury yields.
- Hedge funds got long swap spread when it tightened relative to historical levels, betting that it would widen. Then redemptions and margin calls forced them to unwind these trades forcing the spread to tighten and go negative.
Negative swap spread implies that at some point in the future bank lending rates (LIBOR) will be lower than government yields. This is impossible, given that one can not have a banking system that has better credit than the US government (or is it?). Nevertheless it’s a technical anomaly that is difficult to arb out. It would require purchasing the bond (on repo - in order to get leverage), swapping it into floating, and sitting on the trade for years. It has to be hard to get long-term committed capital these days that is not subject to redemptions.
Having said that, the spread has become significantly less negative - currently at 17 bp. If either of the hypotheses above are correct, any reversal of these trends (institutions buying more fixed rate assets and swapping them into floating or hedge funds putting the widening trade back on) would cause this.