GLOBAL RESEARCH ARCHIVE
US Rates Strategy: Intraday Buyback Effect
Research evidence excerpt
US Rates Strategy: Intraday Buyback Effect
IdeaM
Exhibit 1: Buyback ratio versus offer-to-maximum Exhibit 2: Share of auctioned duration risk bought
multiple by curve sector back through liquidity support operations
Buyback ratio %
16 1.0
0.9 14
0.8
0.7
0.6 10
0.5
0.4
0.3 6
0.2
0.1
0.0 2
0.0 2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0
Offer to Maximum 0
Front-end Belly Back-end TIPS 2Y 3Y 5Y 7Y 10Y 20Y 30Y
Source: US Treasury, Morgan Stanley Research Source: US Treasury, Morgan Stanley Desk estimates, Morgan Stanley Research
Buyback sizes are large in absolute terms, and each operation also tends to be
concentrated in a few CUSIPs.
• This is particularly true beyond the 5y-7y sector, with 10y-20y buybacks the most
concentrated, see Exhibit 3 .
Given that buybacks add demand for meaningful amounts of risk in concentrated off-the-
run CUSIPs, they should affect the liquidity of the bonds involved.
• In the rest of this piece, we investigate how buyback events affect bonds in the
target sector.
° We focus on relative yield performance, as additional demand should reduce
illiquidity premiums, and on bid-ask spreads, as improved liquidity should
lower transaction costs.
• Exhibit 4 shows the timeline of a single buyback operation. The process works as
follows:
° In its quarterly refunding documents, the Treasury releases the buyback
schedule for the upcoming quarter.
° At 11:00am ET on the day before a buyback, the Treasury releases a
preliminary list of CUSIPs within the buyback window that are eligible to be
accepted in the operation.
° At 11:00am ET on the operation date, the Treasury releases a finalized version
of the list, usually with few changes.
° The Treasury then accepts offers from participants in those CUSIPs from
1:40pm to 2:00pm ET on the operation date.
The English excerpt is extracted automatically from the cited source page and may contain layout or recognition errors. It is never batch translated.
Open report viewer