Accessible Version
Treasury Issuance and Dealer Balance Sheets, Accessible Data
Figure 1. Percent of Treasury Security Auction Allocations by Investor Class
Figure 1 has two panels, which both show area graphs that break down the volume of Treasury securities allocated to each investor class, in percentage terms, from 2009 to 2025. The types of investor class are: individuals; depository institutions; pension and retirement funds, insurance companies; dealers and brokers; investment funds; other; and foreign and international. Panel A, which has data for Treasury coupon securities, shows that the share allocated to dealers and brokers has steadily decreased from about 50 percent in 2009 to about 20 percent in 2025. The share allocated to investment funds has steadily increased, from about 10 percent in 2009 to about 50 percent in 2025. Panel B, which has data for Treasury bills, shows a similar pattern. The share of bills allocated to dealers and brokers decreases from about 60 percent in 2009 to about 40 percent in 2025, while the share allocated to investment funds increases from about 10 percent in 2009 to about 50 percent in 2025.
Note: SOMA allocations are excluded from the calculation. Key identifies in order from top to bottom.
Source: U.S. Department of the Treasury.
Figure 2. Dealers' Sensitivity to Coupon Issuance
Figure 2 is a line graph that shows three lines, representing dealers’ positions, repo, and reverse repo, from 2016 to 2025. Each line represents each balance sheet items’ sensitivity to Treasury coupon issuance, as measured by the coefficient of a one-year rolling regression of issuance volumes on each balance sheet item. There are two shaded time periods, denoting the first episode of QT (late 2017 to mid 2019) and the second episode of QT (mid 2022 to mid 2025). The chart shows that the sensitivity of positions has been fairly constant, at around 0.05 to 0.1. Repo and reverse repo tend to increase significantly during the QT periods, particularly the second QT period, when they rise above 0.2.
Note: This chart plots the coefficients of a one−year rolling regression of the daily changes in dealers' positions or repo on Treasury issuance. Shaded areas indicate periods of quantitative tightening.
Source: U.S. Department of the Treasury; Federal Reserve Board, Form FR 2052a, Complex Institution Liquidity Monitoring Report; and staff analysis.
Figure 3. Mean Net Outright Treasury Position
Figure 3 has two panels, which both show line graphs depicting the mean net outright Treasury position in different tenors during an auction cycle, plotted in event time. The different tenors shown are: 2Y, 3Y, 5Y, 7Y, 10Y, 20Y, and 30Y. Panel A has the issuance date centered at zero and shows the mean net outright position for plus and minus ten days around it, while Panel B shows plus and minus ten days centered around the auction date. Both charts show that dealers hold net short positions across all tenors ahead of the auction. On the auction day, net positions spike for all tenors as dealers receive their allocations, with the net position becoming positive for most tenors. In the following days, dealers continue selling through their allocations, with net positions returning near zero within three to five days.
Note: Data is presented as mean gross volumes for all auctions from 2016−03−29 to 2025−07−08, excluding reissued securities.
Source: FR2004, TRACE and Treasury Direct.
Figure 4. Dealer Coupon Inventory Half Life Around Auctions
Figure 4 is a bar chart showing the half-life of dealer coupon inventories around auctions for each tenor. It show that, on average, the half life is close to two days. There is variation across tenors, with the 10-year having a half life under 0.5 days and the 3-year having a half-life close to 3 days.
Source: FR2004, TRACE and Treasury Direct
Figure 5. Supplementary Leverage Ratio
Figure 5 is a line chart that shows the supplementary leverage ratios (SLRs) of 6 large bank holding companies from 2016 to 2025. The 6 companies are: J.P. Morgan Securities LLC; BofA Securities, Inc; Wells Fargo Securities, LLC; Citigroup Global Markets Inc.; Morgan Stanley & Co LLC; and Goldman Sachs & Co LLC. The chart has a shaded area from Q2 2020 to Q1 2021, which indicates the period during which Treasuries and reserves were exempted from the SLR calculation. There is a horizontal line at 5 percent, indicating the minimum level of the SLR allowed for these large bank holding companies. For the period from 2017 through the end of the exemption period in 2021, most of the six firms maintained SLRs well above the 5 percent minimum level. After the exemptions expired, SLRs trended down and, for some firms, have been much closer to the minimum level during 2021 to 2025. The chart also indicates quarters during which a firm’s SLR implied a higher capital requirement than risk-based rules. Morgan Stanley has consistently had higher capital requirements under the SLR from 2021 to the present, while JP Morgan and Goldman Sachs have sometimes had higher capital requirements under the SLR during the same period.
Source: FR Y−9C
Figure 6. Mean VaR Usage for UST Market−Making Desks
Figure 6 is a line chart that shows the mean usage of value-at-risk limits across Treasury market-making desks at large dealers, from 2016 to 2025. One line shows a 30-day moving average of mean VaR usage, while the other line shows the observed usage on each day. The moving average line is generally between 0.3 and 0.5, except during March 2020, at which point it rises to 0.7 before gradually returning to its normal range. The observed usage line is more volatile, and it spikes at 0.9 in March 2020.
Note: VaR usage, calculated at the desk level, is the ratio of a desk's VaR to its daily limit. The figure shows the mean VaR usage across a sample of large dealers' UST market−making desks. The projection is created by a regression over the last twelve months of available data, forecasted to the present.
Source: FR VV−1