Curves have bear flattened since last week. The higher duration, longer-dated treasuries have outperformed the lower duration, shorter-dated treasuries in the selloff.
Wasn't the curve upward sloping? so a flattening would mean the back came in more? I don't really know fixed income
Hasnt the long dated paper % drops been much bigger? sure, in terms of bps it might have been different but in terms of % declines it has been pretty bad, thats the very definition of duration, no?
The curve has been upward sloping, but relatively flat, by historical standards. Falttening in yield terms, in this case of a selloff, meant that the back end yields went up by less than the short end yields. To be fair, not by a lot, but the fact that this has happened after what we have learned is rather amazing.
No, percentages don't tell you the whole story, since you're not taking the duration ratio into account. That's the whole point of using basis points applied to yields, since this assumes that your positions are appropriately weighted. So in this particular case, let's take 5s and 30s, for instance (using the current 5y OTR and the o30s, Aug46s). 5s lost arnd 2.32% in price terms, while 30s lost 9.6%, which means that 30s lost 4.1379 times what the 5s lost. However, the 30s are about 4.3124 times the duration of the 5s.