Developments Affecting Future

Changes in Treasury Auction Sizes

Treasury Borrowing Advisory Committee

February 1, 2022

1
In November 2021, based on projected borrowing needs and consistent with the TBAC’s
recommendations, Treasury began reducing auction sizes across all nominal coupon
securities. How has the market responded to these auction size changes? Looking ahead, what
new developments or additional factors should Treasury consider as it evaluates to what extent
auction sizes should be further changed?

2
Table of Contents
• Executive Summary 4
• Market Response to Reduction in Treasury Auction Sizes 5
• Factors to Consider for Future Auction Size Changes: Fiscal, TGA, SOMA 11
• SOMA: Key Questions for Treasury to Consider 17
• Simulations of Future Treasury Auction Size Changes 23
• Conclusions 30
• Appendix 32

3
Executive Summary
• In response to declining fiscal requirements, Treasury reduced auction sizes across maturities last quarter, with larger cuts
in the 7y and 20y. TBAC’s guidance at the time was that cuts to these issue sizes would be required for a few quarters to
adjust to the expected lower future financing needs.
• The cuts were broadly anticipated by the market. Despite the larger cuts in the 7y and 20y, these maturities still
cheapened, suffering from relatively worse liquidity due to ongoing issuance in excess of real money demand in the
market.
• Ongoing fiscal uncertainty creates a broad range of potential financing outcomes that Treasury must remain poised to
address. When considering future changes to auction sizes, Treasury should take into account recent changes in expected
fiscal requirements, and notably, an increased likelihood that the Fed will initiate runoff of its SOMA portfolio later this
year.
• Our expectation is that the SOMA portfolio runoff will have a significant impact on Treasury’s financing outlook, creating
~$1.6T in new financing needs over the next 3 years, which raises the question of whether the prior TBAC guidance to
keep reducing issue sizes is still appropriate, or whether that guidance should be updated.
• We consider multiple scenarios for how Treasury could move forward with future changes in auction sizes. The scenarios
that seem most appealing continue with the pace of cuts in the current quarter only but then either slow or stop the cuts
thereafter. The scenario that extends cuts past the May refunding are intended to allow Treasury to reduce the longer
end, and particularly 7s and 20s, by more than otherwise.

4
Market Response to Reduction in Treasury
Auction Sizes

5
Market Response to Reduction in Treasury Auction Sizes
Primary dealer expectations prior to Nov
Treasury Refunding announcement
• In its Nov-21 letter to Treasury, TBAC said “It was expected that,
based on current fiscal and economic projections, cuts to these Issue Primary % voting Actual
issues need to continue for a few quarters in order to maintain T-bill dealer with Nov cut
sizes in the recommended range of 15 to 20% of total debt consensus consensus size
outstanding over time.” 2Y -2 91% -2
• Consistent with that guidance, in the Nov-21 Quarterly Refunding 3Y -2 91% -2
Statement, Treasury announced auction size reductions across all
nominal coupons, including the 2y, 3y, and 5y by $2B per month and 5Y -2 91% -2
the 7y by $3B per month. Treasury also announced decreases of $2B
to both the new and reopened 10y and 30y, and decreases of $4B to 7Y -3 91% -3
both the new and reopened 20y.
10Y -3 62% -2
• These reductions were broadly in line with primary dealer and
market expectations. To the right is a table summarizing 20Y -3 43% (T) -4
expectations from 23 primary dealers. For 7y and shorter maturities, 20Y -4 43% (T) -4
the dealer consensus was in line with the actual Treasury decision.
For 10y and longer maturities, there was more divergence of 30Y -2 70% -2
opinion, but forecasts were still very much aligned with the
direction and size of Treasury’s actual decision.

6
Source: Bloomberg article 11/2/2021, “Wall Street Calls Diverge on Cuts for 10- to 30-Year Treasuries,” by Elizabeth Stanton
Market Response to Reduction in Treasury Auction Sizes
• The weeks leading up to and following the Nov 3rd refunding announcement were marked by greater than
normal volatility in US and other DM rates markets, particularly volatility of relative value relationships.
• A combination of both US and international macro and policy factors were the likely catalysts, with de-
risking having the biggest impact on trades with more crowded positioning.
• Anticipation of Treasury’s issuance cuts should have caused the 7y and 20y maturities to outperform,
but despite that, volatility remained high and these issues ultimately cheapened further.
• Ceteris paribus, a reduction in aggregate Treasury supply should lead to a widening of swap spreads, as
Treasury collateral becomes relatively scarcer causing Treasury securities to outperform versus swaps; but
investors needed to weigh the relative impact of falling supply versus the Fed’s announcement later that
same day that it would begin tapering its asset purchases.
• Swap spreads did widen across the curve, both in the month leading up to the Treasury announcement, as
well as in the month following. While there are many factors that can impact swap spreads (see Feb-21
TBAC charge), the move wider is consistent with expectations for the market reaction to reduced issuance.

7
Source: JP Morgan DataQuery
 7y and 20y maturities cheapen despite supply
reductions
• Relative value Treasury investors were understood to
hold substantial long positions in both the 7y and 20y,
perhaps in anticipation of the announcement from
Treasury that these issue sizes would be cut more
than other maturities in the Nov-21 refunding
announcement.
• As evidenced in the weighted* FF-OIS ASW butterfly
chart to the right, both the 7y and 20y came under
sharp pressure just prior to the refunding
announcement, snapped back immediately following
the announcement, and then cheapened further,
particularly in the case of the 20y.
• We emphasize that this cheapening occurred despite
the as-expected issue size cuts by Treasury. It’s likely
that performance would have been significantly worse
absent those cuts.
Source: Barclays, and Presenter Calculations

*Butterfly weightings: 5s7s10s 0.5: 1.0: 0.5; 10s20s30s 0.25: 1.0: 0.75

8
7y and 20y maturities cheapen despite supply
reductions
• In the wake of the November refunding
• Simple comparisons are illustrative:
announcement, the 7y and even more so the 20y
1/26/22 5y 7y 10y 20y 30y underperformed, trading at the cheapest levels of
ASW spread 7 2 6 -20 -18 the past 2 years (see FF-OIS ASW below).
• This is indicative of over-supply of these maturities,
• The 7y ASW spread is 4bps lower than both the 5y and
10y, while the 20y is 8bps lower than the arithmetic which in turn can lead to crowded positioning and
average of the 10y and 30y, and is 2bps lower than 30y. further market dislocation.
• While the 20y has a higher yield than the 30y, this
doesn’t prove that the 20y is cheap. For convexity and
other reasons, the 20y point trades at a higher absolute
yield across the US, GBP, and EUR swaps curves.

Swap Curve 20y 30y

1/26/2022
USD 2.04 1.99
EUR 0.64 0.50
GBP 1.06 0.98
9
Source: Barclays, Bloomberg, and Presenter Calculations
 7y and 20y maturities cheapen despite supply
reductions
• The 7y saw a significant increase in issuance over the past year, $841B in total supply from Jan-21 to Jan-22.
SOMA purchases absorbed only 16% of that supply.
• The 20y part of the curve has relied heavily on Fed support, both through purchases of recently issued 20y
coupons, and because the Fed’s purchases in the 22.5y-30y bucket have been concentrated in shorter
maturities, also supporting the 20y part of the curve. In the 18y-25y bucket, there was $330B in new issuance
over the past year, and $171B of Fed purchases, effectively absorbing 52% of that issuance.
• Both 7y and 20y issue sizes should be reduced further relative to other nominal coupons, to better balance
supply with end-investor demand and to facilitate better liquidity and more orderly markets.

Treasury issuance and SOMA purchases, trailing 12m

Bucket Total Out- SOMA SOMA % Issuance SOMA SOMA %

standing Holdings (T12m) Purchases
8y-10y 1,074B 267B 24.9% 575B 153B 26.6%
10y-18y 263B 184B 70.0% 0B 0B n/a
18y-25y 1,746B 785B 45.0% 330B 171B 51.8%
25y-30y 1,282B 280B 21.8% 367B 100B 27.2% 10
Source: Refinitiv, Presenter Calculations
Factors to Consider for Future Auction Size
Changes:
Fiscal Outcomes and SOMA Path

11
Factors to consider for future auction sizes
• Treasury aims to fund the government at the lowest cost to tax payers overtime.
• To do so, Treasury seeks to be a regular and predictable issuer, while maintaining flexibility to meet
changing fiscal requirements and investor demand dynamics.
• Bills as a percentage of marketable debt outstanding should remain in the 15%-20% range, consistent
with previous TBAC guidance.
• Treasury should take into consideration the effects of its issuance on weighted average maturity
(WAM), weighted average duration (WAD), and other metrics of its outstanding debt. WAD can be
measured in two ways:
1. All Treasury debt is treated the same, regardless of whether it held by the public or by SOMA. By
this measure, the WAD as of 12/31/2021 is 5.0.
2. Treasuries held by the Fed are calculated as having the duration of daily-reset FRNs, regardless of
their contractual maturity. This is consistent with the idea that the Treasury and the Fed should
be thought of as a consolidated balance sheet. By this measure the WAD as 12/31/2021 is 3.8.
• Treasury should continue to monitor the Treasury debt market and take actions if needed to support
liquid and well-functioning markets.

12
Factors to consider for future auction sizes: Fiscal
• As always, Treasury’s borrowing needs will be strongly influenced by future fiscal deficits.
• To calculate, we take into account three inputs:
1. CBO baseline budget projections, last updated July-2021, for the period 2022 through 2031.
2. CBO net changes to deficit estimates due to the Infrastructure Investment and Jobs Act, published Aug.
9, 2021.
3. Assumption of no net funding needs from BBB:
• As negotiations continue, there is uncertainty about the size of both gross and net spending that
would result from passage. Public statements in recent months suggest that a successful bill would
likely be approximately deficit neutral.
• While relying on the CBO baseline projections is helpful for a point-in-time modeling exercise, it should be
emphasized that there is considerable uncertainty about the future size of financing needs.
• In the Aug-2021 presentation, TBAC anticipated $1.5T of additional fiscal financing needs in excess of CBO
baseline, 6 months later we estimate only $347B.
• Treasury needs to account for financing future growth of the Treasury General Account (TGA).
• We assume $650B will be the balance as of YE-2022 as well.
• We assume that TGA will grow at the same rate as nominal GDP in future years.
• Data table with actual projections for each year from 2022-2031 is in the appendix.

13
Factors to consider for future auction sizes: SOMA

• Brief review of the SOMA (System Open Market Account) portfolio,

referencing the Feb-2020 TBAC charge entitled “Implications of the SOMA
Portfolio for Treasury Debt Management.”
• SOMA is the portfolio of assets held by the Federal Reserve System obtained through
open market operations.
• The Treasury and the Federal Reserve are independent entities operating with different
mandates. However, decisions that the Fed makes about the SOMA portfolio could affect
the Treasury’s fiscal outcomes and its debt management decisions.
• A portion of SOMA holdings (that held against currency liabilities) can be thought of as
translating Treasury debt into debt financed at a 0% interest cost to the Treasury.
• Most of the remaining SOMA holdings can be thought of as translating Treasury debt
into floating rate notes (FRNs) that are tied to the overnight interest rates set by the Fed,
most importantly the interest rate paid on reserves, or the IOER rate.

14
SOMA: Balance sheet runoff is likely to begin in 2022
• The SOMA portfolio has been a sizable buyer of Treasuries and MBS since the March-2020 liquidity crisis,
initially to prevent market dysfunction and subsequently at a monthly rate of $80B Treasuries and $40B MBS to
support the economy.
• SOMA currently reinvests all maturing Treasury securities by placing bids at Treasury auctions equal in par
amount to the value of holdings maturing on the issue date of the securities being auctioned, allocated
proportionately across those securities by announced offering amount.
• In November, the Fed started to reduce the size of its purchases and in December the Fed announced that it
would cease new purchases in March, 2022.
• It is now the consensus view among both market and FOMC participants that at some point following the Fed’s
first rate hike, the Fed will initiate balance sheet runoff.
• “Almost all participants agreed that it would likely be appropriate to initiate balance sheet runoff at some point after
the first increase in the target range for the federal funds rate.” FOMC minutes, Dec 14-15, 2021 meeting.
• “The Committee expects that reducing the size of the Federal Reserve’s balance sheet will commence after the process
of increasing the target range for the federal funds rate has begun.” Principles for Reducing the Size of the Federal
Reserve’s Balance Sheet
• Market pricing implies that it is likely the Fed will hike rates in March, meeting the condition to consider
initiating balance sheet runoff.
• “At some point perhaps later this year we will start to allow the balance sheet to run off, and that’s just the road to
normalizing policy.” Chair Powell, Senate confirmation hearings, Jan 11, 2022.
• “I think certainly by summer, we can start shrinking the balance sheet.” – FRB Governor Christopher Waller, Bloomberg
TV interview, Jan 14, 2022

15
 SOMA: 2017-2019 experience

• In Sep-2017, the Fed announced that it would begin

normalization of its balance sheet in Oct-2017, by not
reinvesting up to a cap of $6B UST and $4B MBS per month.
• Consistent with the June-2017 addendum to the committee’s
policy normalization principles and plans, the Fed
incremented up those caps by $6B and $4B respectively until
reaching steady-state caps of $30B UST and $20B MBS.
Maturing or paid-down securities in excess of the caps were
reinvested into UST or MBS respectively.
• There was uncertainty about the optimal long term size of the Fed’s balance sheet. To reduce the risk of
overshooting, the Fed announced in March-2019 that starting in May-2019, it would reduce the cap for UST
from $30B to $15B, while maintaining the cap for MBS at $20B.
• In July-2019, the FOMC announced that MBS principal paydowns up to $20B per month would be reinvested
into UST, with any excess above that invested into MBS. It also announced that there would be no further net
reduction of the aggregate holdings in the SOMA portfolio, two months ahead of schedule.
• The size of the Fed’s balance sheet troughed at $3.8T, 18% of GDP in Sep-2019 when dysfunction in the money
markets occurred. It became clear that this was too low a level for the “ample reserves” system, and the Fed
immediately took action to increase the size of its balance sheet by about 10% over ensuing months. 16
SOMA: Key Decisions that Will Affect
Treasury’s Borrowing Needs

17
 SOMA Questions for Treasury to consider
• There is significant uncertainty about the exact parameters the Fed will choose when it initiates balance sheet
runoff this time around.
• Our goal is to make the best inferences possible based on Fed statements and commentary to construct the
most likely scenarios that Treasury should consider.
• We think the most important aspects for Treasury to focus on will be the size of caps placed on monthly runoff
and how long runoff will continue before the Fed determines that normalization of the sizing of its balance
sheet is complete.
• We briefly address how the Fed might choose to runoff its bill and MBS portfolios. These are relatively less
important from Treasury’s perspective, but could still have non-negligible impact on Treasury issuance needs.
• When will balance sheet runoff begin?
• We assume runoff begins in July-2022. The exact timing is not that critical from Treasury’s perspective.
• Will there be a phase-in period to gradually increase runoff and if so, how long will the phase-in last and at what
intervals will runoff be adjusted?
• “Many participants judged that the appropriate pace of balance sheet runoff would likely be faster than it
was during the previous normalization episode.” – Dec-2021 FOMC minutes.
• We assume there will be a phase-in period, but that it will only be 6 months, not 12 months like in 2017.
• We assume that the caps will increment higher in even intervals each month until reaching the maximum
cap level in the 6th month.
• The exact details of the phase-in period are not that critical from Treasury’s perspective.
18
SOMA Questions for Treasury to consider: Caps
• Assuming there are runoff caps, what size will those caps be?
• “Many participants also judged that monthly caps on the runoff of securities could help ensure that
the pace of runoff would be measured and predictable, particularly given the shorter weighted
average maturity of the Federal Reserve’s Treasury security holdings.” –Dec-2021 Fed minutes.
• “…the balance sheet is bigger so the runoff can be faster.” – Chair Powell, Senate confirmation
hearings 1/11/22
• There is uncertainty about what the final size of the caps will be. We evaluate the following three
scenarios:
• Lower bound: $30B UST / $20B MBS – Equal to the caps set in 2017, this is likely too low since
the Fed says runoff should proceed at a faster pace this time.
• Central case: $60B UST / $30B MBS
• Faster pace than 2017, but significantly slower than the $80B/$40B pace at which the Fed
added securities during 2020 and 2021.
• The 2:1 ratio of UST to MBS better reflects the current ratio of the Fed’s portfolio of UST
(ex-bills): MBS, which is ~2.1:1. In 2017 the ratio was ~1.4:1, consistent with the 1.5:1 ratio
the Fed chose then.
• Upper bound: No caps – This is the upper bound that the Fed could possibly achieve and also
seems unlikely. We demonstrate the impact for Treasury to consider and compare to our central
case.
19
SOMA Questions for Treasury to consider: Caps,
central case
• Our central case for runoff caps is that the Fed chooses $60B UST/$30B MBS.
• The Fed wants to move faster than in 2017, suggesting caps above $30B UST/$20B MBS.
• The Fed is unlikely to have runoff be as fast as purchases, i.e. $80B UST/$40B MBS or greater is too fast.
• That leaves a landing zone of $30B for MBS.
• As mentioned previously, the proportional ratio of UST : MBS is now 2:1, implying a $60B UST cap.
• A Jan 21 Bloomberg story shared estimates from 6 primary dealers:

20
Source: Bloomberg News, 1/21/2022, Bond Dealers Are Mapping Out How Fed Will Shrink Balance Sheet, by Elizabeth Stanton
 SOMA Questions for Treasury to consider: “Normal”
• The ultimate goal of reducing the Fed’s balance sheet is “normalization.” Under the ample reserves system, what is a
“Normal” level of the Fed’s balance sheet, which will correspond to the desired “ample” reserves level?
• The Feb-2017 TBAC charge asked the same question. What is normal?
• 2019 showed that 18% of GDP is too low a level for the Fed’s balance sheet relative to GDP; today’s level of 37% of GDP is
clearly too high.
• We estimate a wide range for normal, 20%-25% of GDP. Given that it will take at least 2-3 years for the Fed’s balance sheet
to near this range, we don’t think it is important for Treasury to try to formulate a more precise view than this.
• Once normalization is complete, the Fed will soon thereafter need to start growing SOMA again, in order to maintain a
proportionately appropriate sized balance sheet as nominal GDP grows. In addition, it is likely that the Fed will reinvest
MBS principal paydowns into Treasuries, leading to additional SOMA demand.
• Different market conditions could result in different answers for what’s normal. For example, it’s likely that demand for
currency would fall in a higher interest rate environment, which in turn would reduce the size of assets the Fed needs to
hold against its currency liability.
• Under the assumptions we have described, the Fed will achieve normalization on the date listed in the left table. Total
runoff by the year is shown in the right table.
Runoff totals by calendar year, Central Case (excludes $326B bills)
Cap Structure UST cap MBS cap Normalization
2022 2023 2024 Total
Date
Central Case $60B $30B Oct-2024 UST 203B 627B 490B 1.32T
Lower Bound $30B $20B Sep-2025 MBS 117B 322B 235B 0.67T
Upper Bound No cap No cap Apr-2024 Total 320B 949B 725B 1.99T 21
Source: Presenter calculations
 SOMA Questions for Treasury to consider: Bills & MBS
• It’s likely the Fed will runoff its $326B portfolio of bills. There are two options:
1. Include the bills under the cap for nominal coupon runoff.
2. Allow bills to runoff separately, with or without separate caps on bill runoff. Total MBS principal paydown
reinvestment expected in excess
• For simplicity, we assume that the Fed runs bills off separately, rather than including of the MBS cap.
them in the overall nominal coupon cap.
• Regardless of the Fed’s choice, Treasury should offset by issuing additional bills to the
public. This is a straightforward way for Treasury to offset SOMA’s bill redemptions and
there is ample liquidity available in the system to absorb additional bills.
• Whether or not the Fed reinvests MBS principal paydowns in excess of the MBS cap back into
MBS or into UST is not that consequential for Treasury.
• “Consistent with the previous normalization principles, some participants expressed a
preference for the Federal Reserve's asset holdings to consist primarily of Treasury
securities in the longer run. To achieve such a composition, some participants favored
reinvesting principal from agency MBS into Treasury securities relatively soon or letting
agency MBS run off the balance sheet faster than Treasury securities.” – Dec-2021 FOMC
minutes 2022 is higher due to assumption
that caps phase in.
• We assume that when the Fed ultimately stops running off the SOMA portfolio, it will
reinvest MBS paydowns into UST, to move in the direction of holding only UST in the long
term.
22
Source: JP Morgan research
Simulations of Future Treasury Auction
Size Changes

23
Simulations of future Treasury auction size changes
• We incorporate the previously mentioned assumptions for the fiscal requirements, TGA, and the expected runoff of the
SOMA portfolio.
• All simulations/scenarios make the following additional common assumptions:
• CBO fiscal, IIJA, and BBB fiscal deficit assumptions, and TGA assumptions.
• SOMA assumptions:
• Balance sheet runoff begins July-2022, is phased in over a 6 month window, and stops when SOMA is equal to
23% of nominal GDP.
• The Fed resumes purchases one year after stopping runoff, maintaining SOMA at a constant percentage of
nominal GDP thereafter.
• MBS principal paydowns above the cap are reinvested into MBS until the Fed stops runoff, and thereafter are
reinvested into UST.
• Bills are run off separately from nominal coupons and are replaced by issuing additional bills to the public.
• For each scenario, we evaluate 3 sets of runoff caps:
• Lower bound: $30B UST / $20B MBS
• Central case: $60B UST / $30B MBS
• Upper bound: no caps

24
 Scenario 1: Freeze auction sizes, except for decreases
in 7y/20y
• Plan: Freeze auction sizes at current levels, except for $3B monthly reduction in 7y and $4B quarterly reduction in 20y for the Feb
quarter only.
• Rationale: Treasury should stop cutting auction sizes to ensure it is issuing sufficiently to meet the higher issuance needs of financing
SOMA runoff net of lower fiscal requirements, while further reducing 7y and 20y auction sizes to better match supply and demand of
those maturities.
• Analysis:
• Regular & predictable in that auction sizes are constant, but a change versus prior guidance, i.e. less predictable.
• The bill percentage hovers right at the 15% threshold in the central scenario and falls well below in the lower bound scenario.
• There is a sharp increase in the WAM and WAD of the outstanding debt. This may not be desirable.
• 7y/20y issuance falls on a relative and absolute basis.
• Expected future adjustments: Treasury would likely need to make future cuts in coupon auction sizes to prevent bill percentage from
falling below 15%.

25
26
27
 Scenario 4: Twist, increase front end, decrease long
end issuance, especially 7y/20y
• Plan: In Feb only, make monthly increases in the 2y, 3y, and 5y by $2B, reduce the 7y monthly by $3B, and make quarterly cuts in
the 10y by $3B, the 20y by $4B, and the 30y by $2B.
• Rationale: As Treasury replaces debt previously held by SOMA with future issuance to the public, the effective (treating SOMA
holdings as daily-reset FRNs) WAD of the Treasury’s total issuance will rise. Increasing front end issuance and decreasing long
end issuance will help offset. Combines elements of Freeze and Reduce.
• Analysis:
• To flip from cuts last quarter to increases this quarter is not regular & predictable, only makes sense if a significant change
in circumstance justifies it.
• Bill percentage falls below 15% threshold in central scenario.
• WAM and WAD rise significantly, although there is a partial reversal in 2027-2031.
• 7y/20y issuance falls on a relative and absolute basis.
• Expected future adjustments: Treasury may need to make cuts later in 2022 to prevent breach of 15% bill percentage threshold.

28
Continue or Extend scenarios seem most favorable

• The simulations show that cutting for fewer quarters than suggested by the previous TBAC guidance is now appropriate,
given the change in fiscal requirements and expected SOMA runoff.
• There still should be some cuts, as evidenced by the Freeze scenario, which would result in over-funding, leading to bill
percentage falling below the 15% lower bound, as well as a sharp increase in WAM/WAD.
• Cuts could continue for one additional quarter at the same pace as last quarter, which works well for managing bill
percentage and results in a modest increase in WAM and WAD.
• An alternative is to extend cuts, first by implementing the same cuts as Continue in Feb, and then making additional cuts
in long end issuance in the May quarter. This reduces the increase in WAM/WAD, and results in further cuts to the issue
sizes of the 7y and 20y.
• Attempting to twist issuance by increasing front end issuance and reducing long end issuance results in over-funding,
violating the bill percentage lower bound, necessitating coupon cuts later in 2022 or early in 2023.
29
Conclusions

30
Conclusions
• Treasury’s coupon cuts announced in November were widely anticipated by the market, but they
were not sufficient to resolve the supply/demand imbalance in the 7y and 20y maturities, which
cheapened further despite Treasury’s effort to address.
• Treasury seeks to be a regular and predictable issuer, while maintaining flexibility to meet
changing fiscal requirements and investor demand dynamics.
• In considering future coupon sizing, Treasury should take into account recent shifts in expected
budget deficits resulting from fiscal policy and an increased likelihood of the Fed initiating a
balance sheet reduction of its SOMA portfolio.
• As it relates to the SOMA portfolio, the Fed will likely begin reductions this summer, at a faster
pace than in 2017-2019, but still using caps to maintain limits on the size of monthly reductions.
• To address the various concerns Treasury faces in making future auction decisions, we advise that
the Treasury make further cuts in its issuance in Feb equivalent to the cuts made in the prior
quarter, and consider making additional cuts in the longer end of the curve in May.
• Making additional cuts as suggested should keep the bill percentage in the TBAC suggested 15%-
20% range, result in only modest increases in the WAM/WAD of total Treasury debt issuance, and
further address the supply/demand imbalance in the 7y and 20y issues.

31
Appendix

32
Appendix: Fiscal deficit and TGA data tables, 2022-
2031
Year CBO IIJA BBB Total Year Fiscal TGA Total
2022 -1380 -5 0 -1385 2022 -1385 -244 -1629
2023 -764 -18 0 -782 2023 -782 -28 -810
2024 -803 -38 0 -841 2024 -841 -22 -863
2025 -1008 -47 0 -1055 2025 -1055 -24 -1079
2026 -1095 -69 0 -1164 2026 -1164 -26 -1190
2027 -1123 -63 0 -1186 2027 -1186 -29 -1215
2028 -1548 -48 0 -1596 2028 -1596 -29 -1626
2029 -1386 -32 0 -1418 2029 -1416 -30 -1448
2030 -1794 -25 0 -1819 2030 -1819 -31 -1850
2031 -1914 -2 0 -1916 2031 -1916 -33 -1949

33