While the minutes for the September MPC meeting clearly indicated the risks were to the upside the timing and size came as a surprise.
Purchases started today and the range of gilt purchases is unchanged at three years-plus, and will be spread evenly across the sectors. It is expected to take four months to complete, arguably given the MPC the opportunity to review it at the February meeting.
This compares to remaining gross issuance over the rest of the fiscal year of £75.9 bn, of which £56.2bn is in nominal gilts. It also overwhelms net issuance which amounts to £60.2bn.
The additional £75bn of purchases is the equivalent of around 5.2% of 2010 GDP, and combined with the first round equates to 18.9% of 2010 GDP – somewhat more than the 16% of 2010 GDP the Federal Reserve has done so far.
The current round is already benefitting the long end relatively more than in 2009, where the range was only expanded partially through the programme. At completion, QE1 was composed of around 13% of 25 year-plus bonds, unlike the equal split seen this time around.
The Bank of England will keep its cap on holdings at 70% of the free float, same as in 2009. Currently the 8% 21 June are excluded as they already fall above this range. In addition, bonds which are being auctioned within a week of purchase are excluded.
We review the cross market trades, given the outperformance seen over the week. With QE having been delivered in the UK, and already occurring in the US, we see less scope for outperformance versus the US and close the long sterling 5Y5Y versus dollar-denominated 5Y5Y.
However we maintain the long position versus France in the 30-year and ten-year sectors.
This round of purchases should be relatively more supportive of long-end gilts, and should lead to further swap spread widening. In addition, there is room for the 10Y30Y curve to flatten further.
Similarly, we maintain a constructive stance on ten-year breakevens in the UK, which despite having widened last week, remain at attractive levels.
