Overview

This report examines the US treasury security auction market. Each week the US government issues debt in US$ billions in bills, notes and bonds to investors by way of auctions. Details on each auction can be accessed on the Treasury website(U.S. Department of the Treasury, 2025). For each there is typically the announcement and the result. The auction is comprised of competitive offers and non competitive offers. The non competitive portion is priced off the competitive result.

Our interest in the subject was spurred on by anecdotal evidence that suggests a link between the bond market and the US government decision to suspend tariffs. And so we have set out a broad framework that we hope to use to study future treasury auction performance.

The US government uses funds raised from the auction market to: finance its operations (e.g. for spending that exceeds tax revenue); manage its debt (by replacing existing debt or issuing fresh debt); and, to influence monetary policy (in the yield demanded by investors which serves as a benchmark for pricing many other assets).

We present here various statistics for the different securities types issued since 2018. And structure the analysis around the quantum of issuance, level of demand, and, buyer type. From that we derive broad benchmarks that can serve as reference points to evaluate upcoming auction performance. We have examined in more detail the 26 week bill which is among the most actively issued us government security. And after that we examined closely the recent auction of 17 week bills from the standpoint of bid to cover and buyer allocation. Unfortunately the result was inconclusive with no discernible change in buyer allocation but a significant reduction in demand for the issue (manifest in low bid to cover ratio compared to history). Further analysis is required to examine that ratio more fully.

Report method

Our data set is comprised of past auction results from the years 2018 to 10th April 2025. Data was downloaded by python in the form of JSOn file and imported into R. It was then formatted so that field data matched its characteristic form (numeric, factors, characters). Analysis began with a summary of the various security types which is shown below.

Analysis

Debt Type

Since 2018 we find there has been 2940 securities issued for $140 trillion. Of these bills (terms less than 1 year) represent the most sizable issuance type of around 80% ($112tn) of total issuance. Coming next are notes (1-10 year term) that contribute 18% or $25tn and bonds of $3tn.

Security Type	Count	Total Accepted (Billion)	Percent of Total Accepted (%)
Summary of Total Accepted by Security Type (2018-2025)
Bill	2174	112451	80
Note	603	24675	18
Bond	165	3328	2
Total	2942	140454	100

Debt by Year

The tables below set out the count and value of issuance by year for different types of debt. While the count of issues has grown by less than 2x over the six year period issuance value has grown by much more.

Security Type	2018	2019	2020	2021	2022	2023	2024	2025	Total (Billion USD)
Treasury Records Summary by Year (Total Accepted in USD Billion)
Bill	7806	9136	17056	14373	12904	19182	24663	7333	112453
Note	2470	2684	3386	4419	3311	3065	4171	1171	24677
Bond	215	252	510	721	516	453	500	161	3328
Total	10491	12072	20952	19513	16731	22700	29334	8665	140458

Insight: Debt issuance by the US government has accelerated sharply in the past six years from US$10tn in 2018 to US$29tn in 2024 and could reach $34tn for 2025.

Security Type	2018	2019	2020	2021	2022	2023	2024	2025	Total Count
Treasury Records Summary by Year (Count)
Bill	183	229	397	337	275	320	332	101	2174
Note	84	84	83	82	80	82	85	23	603
Bond	15	14	22	27	25	26	27	9	165
Total	282	327	502	446	380	428	444	133	2942

Insight: The count of debt issued has risen from 282 in 2018 to 444 in 2024 and is on course for 531 for 2025.

Debt by Term

Within these broad categories (bills, notes and bonds) securities are issued with different duration. An indication of that is seen in the table which presents debt securities from the shortest term (1 day) to the longest term (30 years) with total issuance and other aggregated metrics (such as the minimum, maximum, average size of each issue).

Security Term	Bill Count	Note Count	Bond Count	Total Accepted (Billion)	Minimum Accepted (Billion)	Maximum Accepted (Billion)	Average Accepted (Billion)	Median Accepted (Billion)	Average Days
Treasury Records Summary (2018-2025-04-09)
1-Day	1	0	0	15	15	15	15	15	1
3-Day	1	0	0	25	25	25	25	25	3
4-Day	1	0	0	0	0	0	0	0	4
6-Day	1	0	0	0	0	0	0	0	6
7-Day	2	0	0	30	0	30	15	15	7
8-Day	3	0	0	65	0	40	22	25	8
12-Day	2	0	0	65	25	40	33	33	12
14-Day	4	0	0	220	50	60	55	55	14
16-Day	4	0	0	105	15	40	26	25	16
17-Day	2	0	0	90	45	45	45	45	17
18-Day	1	0	0	50	50	50	50	50	18
21-Day	7	0	0	295	30	60	42	40	21
22-Day	1	0	0	40	40	40	40	40	22
23-Day	1	0	0	60	60	60	60	60	23
27-Day	1	0	0	60	60	60	60	60	27
4-Week	381	0	0	21413	11	96	56	51	28
30-Day	1	0	0	40	40	40	40	40	30
33-Day	1	0	0	65	65	65	65	65	33
35-Day	2	0	0	100	40	60	50	50	35
37-Day	2	0	0	80	20	60	40	40	37
38-Day	1	0	0	50	50	50	50	50	38
39-Day	2	0	0	105	45	60	53	53	39
40-Day	3	0	0	155	45	60	52	50	40
41-Day	5	0	0	273	23	80	55	65	41
42-Day	150	0	0	8120	20	85	54	60	42
6-Week	9	0	0	593	70	84	74	72	42
43-Day	8	0	0	500	30	80	63	65	43
44-Day	2	0	0	110	50	60	55	55	44
45-Day	1	0	0	35	35	35	35	35	45
48-Day	1	0	0	40	40	40	40	40	48
55-Day	1	0	0	50	50	50	50	50	55
8-Week	340	0	0	17754	25	90	52	43	56
56-Day	1	0	0	30	30	30	30	30	56
57-Day	1	0	0	50	50	50	50	50	57
64-Day	1	0	0	40	40	40	40	40	64
67-Day	1	0	0	40	40	40	40	40	67
69-Day	3	0	0	125	35	45	42	45	69
70-Day	1	0	0	30	30	30	30	30	70
78-Day	1	0	0	60	60	60	60	60	78
79-Day	1	0	0	20	20	20	20	20	79
13-Week	381	0	0	23056	36	92	61	61	91
102-Day	1	0	0	40	40	40	40	40	102
103-Day	4	0	0	130	30	40	33	30	103
105-Day	42	0	0	1140	25	40	27	25	105
118-Day	1	0	0	30	30	30	30	30	118
119-Day	131	0	0	4298	25	40	33	30	119
17-Week	130	0	0	6853	34	64	53	57	119
120-Day	1	0	0	30	30	30	30	30	120
149-Day	1	0	0	45	45	45	45	45	149
154-Day	47	0	0	1515	30	45	32	30	154
156-Day	1	0	0	45	45	45	45	45	156
161-Day	1	0	0	50	50	50	50	50	161
26-Week	381	0	0	20658	36	79	54	54	182
273-Day	4	0	0	90	20	25	23	23	273
52-Week	96	0	0	3472	20	53	37	38	364
1-Year 10-Month	0	30	0	652	0	28	22	22	672
1-Year 11-Month	0	30	0	655	0	31	22	22	703
2-Year	0	117	0	5433	17	76	46	44	730
3-Year	0	88	0	4653	25	89	53	53	1095
4-Year 4-Month	0	1	0	14	14	14	14	14	1566
4-Year 8-Month	0	1	0	15	15	15	15	15	1688
4-Year 10-Month	0	13	0	224	0	22	17	19	1749
5-Year	0	102	0	5079	17	77	50	48	1824
7-Year	0	88	0	3943	31	75	45	41	2556
9-Year 8-Month	0	14	0	203	12	18	14	14	3516
9-Year 10-Month	0	45	0	1230	11	48	27	24	3587
9-Year 11-Month	0	30	0	953	0	42	32	33	3622
10-Year	0	44	0	1622	13	63	37	39	3647
19-Year 10-Month	0	0	20	337	12	29	18	14	7228
19-Year 11-Month	0	0	21	352	0	28	17	13	7260
20-Year	0	0	20	431	16	33	22	18	7290
29-Year 4-Month	0	0	1	5	5	5	5	5	10699
29-Year 6-Month	0	0	6	49	7	9	8	8	10761
29-Year 8-Month	0	0	1	5	5	5	5	5	10823
29-Year 10-Month	0	0	30	613	12	28	20	20	10896
29-Year 11-Month	0	0	29	578	13	27	20	20	10927
30-Year	0	0	37	957	8	41	26	28	10954
Total	2174	603	165	140453	0	96	37	40	NA

Insight: The majority of the US$140 trillion invested in 2942 securities[^1] are bills with a term of less than a year followed by notes (1-10 year) and then bonds (over 10 years).

Debt Details

In the following section are tables that examine what share of issuance is from reopened or newly issued securities. And then tables that exclusively look at Bills, Notes and Bonds.

Reopening

Among the classifications treasury uses to distinguish securities is whether they are fresh newly issued securities at auction or are reopened securities. Grok speculates that the reason for opening pre existing securities for fresh capital could be for operational, market and debt management reasons. For example when a 10 year note issued two months ago may be reopened which results in a fresh note of 9 years and 10 months. That increases the outstanding amount for that specific security while avoiding the need to create a new issue.

Security Term	Accepted (Reopened)	Accepted (Not Reopened)	Average Days (no)
Reopened Securities Summary (in Billions)
1-Day	15	0	1
3-Day	0	25	3
4-Day	0	0	4
6-Day	0	0	6
7-Day	30	0	7
8-Day	65	0	8
12-Day	65	0	12
14-Day	220	0	14
16-Day	105	0	16
17-Day	45	45	17
18-Day	0	50	18
21-Day	295	0	21
22-Day	40	0	22
23-Day	60	0	23
27-Day	60	0	27
4-Week	21413	0	28
30-Day	40	0	30
33-Day	65	0	33
35-Day	60	40	35
37-Day	80	0	37
38-Day	50	0	38
39-Day	105	0	39
40-Day	155	0	40
41-Day	273	0	41
42-Day	8120	0	42
6-Week	593	0	42
43-Day	500	0	43
44-Day	110	0	44
45-Day	0	35	45
48-Day	0	40	48
55-Day	50	0	55
56-Day	30	0	56
8-Week	14739	3015	56
57-Day	0	50	57
64-Day	40	0	64
67-Day	0	40	67
69-Day	0	125	69
70-Day	30	0	70
78-Day	60	0	78
79-Day	20	0	79
13-Week	23056	0	91
102-Day	0	40	102
103-Day	0	130	103
105-Day	1045	95	105
118-Day	30	0	118
119-Day	1560	2738	119
17-Week	0	6853	119
120-Day	30	0	120
149-Day	45	0	149
154-Day	45	1470	154
156-Day	45	0	156
161-Day	50	0	161
26-Week	5166	15492	182
273-Day	90	0	273
52-Week	0	3472	364
1-Year 10-Month	652	0	672
1-Year 11-Month	655	0	703
2-Year	81	5353	730
3-Year	56	4597	1095
4-Year 4-Month	14	0	1566
4-Year 8-Month	15	0	1688
4-Year 10-Month	224	0	1749
5-Year	265	4814	1824
7-Year	0	3943	2556
9-Year 8-Month	203	0	3516
9-Year 10-Month	1230	0	3587
9-Year 11-Month	953	0	3622
10-Year	0	1622	3647
19-Year 10-Month	337	0	7228
19-Year 11-Month	352	0	7260
20-Year	0	431	7290
29-Year 4-Month	5	0	10699
29-Year 6-Month	49	0	10761
29-Year 8-Month	5	0	10823
29-Year 10-Month	613	0	10896
29-Year 11-Month	578	0	10927
30-Year	0	957	10954
Total	84982	55472	1527

Insight: Some 60% of dollars invested ($84tn) is for reopened securities. Of those reopened securities 85% is from five Bills with terms: 13 weeks ($23tn); 4 week ($21tn); 8 week ($14tn); 42 days ($8tn); and, 26 weeks ($5tn). For freshly issued securities ($55tn) a combination of 8 notes and bills make up 80% of invested dollars: 26 week ($15tn); 17 week ($7tn); 2 year ($5tn); 5 year ($5tn); 3 year ($4tn); 7 year ($4tn); 52 week ($3tn); 8 week ($3tn)

Bill issuance

Security Term	Bill Count	Total Accepted (Billion)	Minimum Accepted (Billion)	Maximum Accepted (Billion)	Average Accepted (Billion)	Median Accepted (Billion)	Avg Duration (Days)
Bills Only Summary - Treasury Records
13-Week	381	23056	36	92	61	61	91
4-Week	381	21413	11	96	56	51	28
26-Week	381	20658	36	79	54	54	182
8-Week	340	17754	25	90	52	43	56
42-Day	150	8120	20	85	54	60	42
17-Week	130	6853	34	64	53	57	119
119-Day	131	4298	25	40	33	30	119
52-Week	96	3472	20	53	37	38	364
154-Day	47	1515	30	45	32	30	154
105-Day	42	1140	25	40	27	25	105
6-Week	9	593	70	84	74	72	42
43-Day	8	500	30	80	63	65	43
21-Day	7	295	30	60	42	40	21
41-Day	5	273	23	80	55	65	41
14-Day	4	220	50	60	55	55	14
40-Day	3	155	45	60	52	50	40
103-Day	4	130	30	40	33	30	103
69-Day	3	125	35	45	42	45	69
44-Day	2	110	50	60	55	55	44
16-Day	4	105	15	40	26	25	16
39-Day	2	105	45	60	53	53	39
35-Day	2	100	40	60	50	50	35
17-Day	2	90	45	45	45	45	17
273-Day	4	90	20	25	23	23	273
37-Day	2	80	20	60	40	40	37
12-Day	2	65	25	40	33	33	12
33-Day	1	65	65	65	65	65	33
8-Day	3	65	0	40	22	25	8
23-Day	1	60	60	60	60	60	23
27-Day	1	60	60	60	60	60	27
78-Day	1	60	60	60	60	60	78
161-Day	1	50	50	50	50	50	161
18-Day	1	50	50	50	50	50	18
38-Day	1	50	50	50	50	50	38
55-Day	1	50	50	50	50	50	55
57-Day	1	50	50	50	50	50	57
149-Day	1	45	45	45	45	45	149
156-Day	1	45	45	45	45	45	156
102-Day	1	40	40	40	40	40	102
22-Day	1	40	40	40	40	40	22
30-Day	1	40	40	40	40	40	30
48-Day	1	40	40	40	40	40	48
64-Day	1	40	40	40	40	40	64
67-Day	1	40	40	40	40	40	67
45-Day	1	35	35	35	35	35	45
118-Day	1	30	30	30	30	30	118
120-Day	1	30	30	30	30	30	120
56-Day	1	30	30	30	30	30	56
7-Day	2	30	0	30	15	15	7
70-Day	1	30	30	30	30	30	70
3-Day	1	25	25	25	25	25	3
79-Day	1	20	20	20	20	20	79
1-Day	1	15	15	15	15	15	1
4-Day	1	0	0	0	0	0	4
6-Day	1	0	0	0	0	0	6
Total	2174	112450	0	96	41	40	68

Insight: 80% of dollars invested ($112tn) is for Bills. The top 5 are: 13 weeks ($23tn); 4 week ($21tn); 26 weeks ($21tn); 8 weeks ($18tn); and, 42 day ($8tn). The broad single issuance range for these is between $11-96bn with the median range between $43-61 bn.

Note issuance

Security Term	Note Count	Total Accepted (Billion)	Minimum Accepted (Billion)	Maximum Accepted (Billion)	Average Accepted (Billion)	Median Accepted (Billion)	Avg Duration (Days)
Notes Only Summary - Treasury Records
2-Year	117	5433	17	76	46	44	730
5-Year	102	5079	17	77	50	48	1824
3-Year	88	4653	25	89	53	53	1095
7-Year	88	3943	31	75	45	41	2556
10-Year	44	1622	13	63	37	39	3647
9-Year 10-Month	45	1230	11	48	27	24	3587
9-Year 11-Month	30	953	0	42	32	33	3622
1-Year 11-Month	30	655	0	31	22	22	703
1-Year 10-Month	30	652	0	28	22	22	672
4-Year 10-Month	13	224	0	22	17	19	1749
9-Year 8-Month	14	203	12	18	14	14	3516
4-Year 8-Month	1	15	15	15	15	15	1688
4-Year 4-Month	1	14	14	14	14	14	1566
Total	603	24676	0	89	30	24	2073

Insight: 18% of total dollars invested is allocated to Notes. The top security terms (by total accepted) are: 2 year note ($5tn); 5 year ($5tn); 3 year ($4.6tn); 7 year ($4tn); and, 10 year and those originally issued as ten year ($4tn). Single issuance values range between $11-90 bn. While median range is between $24-53 bn reflecting somewhat smaller average issuance sizes as terms extend.

Bond issuance

Security Term	Bond Count	Total Accepted (Billion)	Minimum Accepted (Billion)	Maximum Accepted (Billion)	Average Accepted (Billion)	Median Accepted (Billion)	Avg Duration (Days)
Bonds Only Summary - Treasury Records
30-Year	37	957	8	41	26	28	10954
29-Year 10-Month	30	613	12	28	20	20	10896
29-Year 11-Month	29	578	13	27	20	20	10927
20-Year	20	431	16	33	22	18	7290
19-Year 11-Month	21	352	0	28	17	13	7260
19-Year 10-Month	20	337	12	29	18	14	7228
29-Year 6-Month	6	49	7	9	8	8	10761
29-Year 4-Month	1	5	5	5	5	5	10699
29-Year 8-Month	1	5	5	5	5	5	10823
Total	165	3327	0	41	16	14	9649

Insight: Approximately 2% of total dollars invested is allocated to bonds with the greatest allocation to 30 year. The range of issue size is between $8-40 billion. And median size is between $20-28 billion.

Other Debt types

Within the categories of Bill, Note and Bond there are cash management bills within bills, floating rate notes, and, treasury inflation protected securities that are issued as notes and bonds.

Cash management bills

Grok suggests the purpose for cash management bills is to smooth out uneven cash flows such as tax receipts or large expenditures. The bills have a less than one year duration and between 2018-2025 there have been $19tn issued representing 16% of total bills and 13% of total issued securities.

Security Term	Bill Count	Total Accepted (Billion)	Minimum Accepted (Billion)	Maximum Accepted (Billion)	Average Accepted (Billion)	Median Accepted (Billion)	Average Days
Treasury Records Summary for CMB (2018-2025-04-09)
42-Day	150	8120	20	85	54	60	42
119-Day	131	4298	25	40	33	30	119
154-Day	47	1515	30	45	32	30	154
105-Day	42	1140	25	40	27	25	105
43-Day	8	500	30	80	63	65	43
21-Day	7	295	30	60	42	40	21
41-Day	5	273	23	80	55	65	41
14-Day	4	220	50	60	55	55	14
40-Day	3	155	45	60	52	50	40
103-Day	4	130	30	40	33	30	103
69-Day	3	125	35	45	42	45	69
44-Day	2	110	50	60	55	55	44
16-Day	4	105	15	40	26	25	16
39-Day	2	105	45	60	53	53	39
35-Day	2	100	40	60	50	50	35
273-Day	4	90	20	25	23	23	273
17-Day	2	90	45	45	45	45	17
37-Day	2	80	20	60	40	40	37
8-Day	3	65	0	40	22	25	8
12-Day	2	65	25	40	33	33	12
33-Day	1	65	65	65	65	65	33
27-Day	1	60	60	60	60	60	27
78-Day	1	60	60	60	60	60	78
23-Day	1	60	60	60	60	60	23
38-Day	1	50	50	50	50	50	38
55-Day	1	50	50	50	50	50	55
57-Day	1	50	50	50	50	50	57
18-Day	1	50	50	50	50	50	18
161-Day	1	50	50	50	50	50	161
156-Day	1	45	45	45	45	45	156
149-Day	1	45	45	45	45	45	149
64-Day	1	40	40	40	40	40	64
30-Day	1	40	40	40	40	40	30
48-Day	1	40	40	40	40	40	48
102-Day	1	40	40	40	40	40	102
67-Day	1	40	40	40	40	40	67
22-Day	1	40	40	40	40	40	22
45-Day	1	35	35	35	35	35	45
7-Day	2	30	0	30	15	15	7
70-Day	1	30	30	30	30	30	70
120-Day	1	30	30	30	30	30	120
118-Day	1	30	30	30	30	30	118
56-Day	1	30	30	30	30	30	56
3-Day	1	25	25	25	25	25	3
79-Day	1	20	20	20	20	20	79
1-Day	1	15	15	15	15	15	1
4-Day	1	0	0	0	0	0	4
6-Day	1	0	0	0	0	0	6
Total	456	18651	0	85	39	40	NA

Insight: During the period some 456 cash management bills were issued for $19tn. The top ranked durations were 42 days ($8tn); 119 days ($4tn); 154 days ($1.5tn); 105 days ($1.1tn); and, 43 days ($0.5tn). The median issue size for these bills is between $25-60billion.

FRN

A separate category of security issued as notes are floating rate notes (FRN). In this security interest payments fluctuate as interest rates change. And so as interest rates rise the security’s rate will rise and offer higher returns.

Security Term	Note Count	Total Accepted (Billion)	Minimum Accepted (Billion)	Maximum Accepted (Billion)	Average Accepted (Billion)	Median Accepted (Billion)	Average Days
Treasury Records Summary for FRN (2018-2025-04-09)
2-Year	29	745	17	33	26	26	730
1-Year 11-Month	30	655	0	31	22	22	703
1-Year 10-Month	30	652	0	28	22	22	672
Total	89	2052	0	33	23	22	NA

Insight: Relative to the scale of other treasury securities FRNs are a small percentage of total issuance (1%). Two years is the common duration. And the issue size ranges between $22-26 billion.

TIPs

Treasury Inflation Protected securities (TIPS) aim to provide protection against rising inflation. Features of these securities is that the principal value adjusts based on changes in the consumer price index (principal increases with inflation and decreases with deflation). Interest payments are calculated on the adjusted principal. TIPS are issued as notes for the most part and bonds and represent a small percentage of total securities issued by the government (less than 1% of total).

Security Term	Note Count	Bond Count	Total Accepted (Billion)	Minimum Accepted (Billion)	Maximum Accepted (Billion)	Average Accepted (Billion)	Median Accepted (Billion)	Average Days
Treasury Records Summary for TIPS (2018-2025-04-09)
5-Year	14	0	261	17	24	20	20	1811
10-Year	15	0	247	13	21	16	16	3637
4-Year 10-Month	13	0	224	0	22	17	19	1749
9-Year 10-Month	15	0	218	11	19	15	15	3577
9-Year 8-Month	14	0	203	12	18	14	14	3516
30-Year	0	8	75	8	11	9	9	10945
29-Year 6-Month	0	6	49	7	9	8	8	10761
4-Year 8-Month	1	0	15	15	15	15	15	1688
4-Year 4-Month	1	0	14	14	14	14	14	1566
29-Year 4-Month	0	1	5	5	5	5	5	10699
29-Year 8-Month	0	1	5	5	5	5	5	10823
Total	73	16	1316	0	24	13	14	NA

Insight: the majority of TIPS are issued with a 5 year and 10 year term. And the median size for each issue ranges between $14-20 billion.

26 week bill

The 26 week bill is among the most frequent issued security. Since 2018 there have been $15tn fresh issued bills and $5tn reopened. We collate the securities by month and plot the trends.

Month	Count 2018	Count 2019	Count 2020	Count 2021	Count 2022	Count 2023	Count 2024	Count 2025	Accepted 2018 (Billion)	Accepted 2019 (Billion)	Accepted 2020 (Billion)	Accepted 2021 (Billion)	Accepted 2022 (Billion)	Accepted 2023 (Billion)	Accepted 2024 (Billion)	Accepted 2025 (Billion)
26-Week Treasury Bills (Freshly Issued): Frequency and Total Accepted by Month and Year (2018-2025-04-09)
January	3	4	4	3	3	3	3	4	126	150	149	174	172	157	218	307
February	3	3	3	3	3	3	4	3	129	117	119	172	172	162	304	223
March	4	3	3	3	4	4	3	3	180	117	119	172	219	207	211	206
April	3	3	4	4	3	3	3	2	126	111	200	241	159	160	223	72
May	4	4	3	3	3	3	3	0	168	144	179	179	144	171	229	0
June	3	3	3	3	3	3	3	0	126	108	174	174	138	182	213	0
July	3	3	3	3	3	3	3	0	132	108	173	176	144	188	222	0
August	3	3	3	3	3	4	4	0	135	123	171	170	143	266	304	0
September	3	3	3	4	4	3	3	0	126	126	169	207	185	192	213	0
October	3	4	4	3	3	3	4	0	123	168	226	151	147	211	308	0
November	4	3	3	3	3	4	3	0	153	124	171	165	153	296	232	0
December	3	3	4	4	4	3	3	0	108	109	226	225	197	208	218	0
Total	39	39	40	39	39	39	39	12	1632	1505	2076	2206	1973	2400	2895	808

Insight: There are around 3 issues of the 26 week bill per month. The monthly amount raised has varied between $109-300billion since 2018.

##  Date[1:88], format: "2018-01-01" "2018-02-01" "2018-03-01" "2018-04-01" "2018-05-01" ...

Insight: With the monthly issuance series decomposed it appears as though there is some seasonality in issuance. There is also a distinct steady rising trend in the value of issuance since 2018.

Insight: The box plot shows the increasing value of 26 week securities by year with record issuance in 2024. Red dots indicate outlier values for certain issues.

17 week bill

Was the result of the 17-week bill auction on 9th a factor in causing a deferral of US tariffs? To assess this we examine two measures: the bid to cover ratio; and, the composition of buyer type for the bills.

Over the past six years there has been 130 issues of the 17 week bill that has raised $7tn. The average size for each issue was around $50bn with a range of between $34-64 billion. The auction of the 9th was at the high end of that range at $60 billion. The auction saw a bid to cover ratio of 2.71 that was based on $162bn tendered and $60bn accepted.

We look now at how these metrics have performed over time and against that see how the 9th auction compared.

Bid to cover

The bid to cover ratio is a measure of investor appetite for an issue. It takes the amount tendered for the security and compares that number to the total accepted amount. If there is strong interest the ratio should be above average. While conversely if there is weaker sentiment the ratio would be below past trends.

Insight: The auction result showed a lower than average bid to cover ratio at 2.71 when compared to the median and mean for the series of 3.0. That suggests the bond attracted weaker than usual demand.

Buyer type

To examine whether there was a change in the pattern of demand for these bills we examine the allocation of the bills relative to history. For this purpose we have distinguished between competitive bids (by primary dealers, direct bidders and indirect bidders that include foreign and international monetary authorities who place bids through a direct submitter) which are used to set the price for the auction, non competitive bids (that include awards to treasury retail, foreign and international monetary authorities and other non competitive bidders), and, the system open market account (SOMA) which represents the federal reserve.

Descriptive statistics

Investor	Average Allocation
Average Allocation by Investor
17-Week Bill Auctions (Historical Data)
Primary Dealer (own house account)	0.387
Direct Bidder (own house account)	0.045
Indirect Bidder (including foreign & intl monetary authorities)	0.543
FIMA Noncompetitive (foreign & intl monetary authorities)	0.004
Treasury Retail Noncompetitive	0.008
Other Noncompetitive	0.004
SOMA (managed by Federal Reserve)	0.009
Total	1.000

Insight: Based on the history of auctions for the 17 week bills since 2018 most bidders are indirect with an allocation of 54%. Primary dealers are the next significant bidder taking up 39% of issues in the past.

Investor	Standard Deviation
Allocation Variability by Investor
17-Week Bill Auctions (Historical Data)
Primary Dealer (own house account)	0.061
Direct Bidder (own house account)	0.023
Indirect Bidder (including foreign & intl monetary authorities)	0.067
FIMA Noncompetitive (foreign & intl monetary authorities)	0.010
Treasury Retail	0.002
Other Noncompetitive	0.001
SOMA (managed by Federal Reserve)	0.009

Insight: Indirect bidders show highest variability in allocation (6.7%), indicating significant fluctuation in their participation. Primary dealers show less variability (6.1%). Direct bidders have relatively low variability (2.3%), indicating stable participation. FIMA noncompetitive, Treasury Retail, Other Noncompetitive, and SOMA allocations are minimal, with low variability, indicating stable but small participation.

Investor	Median	Q1	Q3	IQR
Central Tendency and Spread by Investor
17-Week Bill Auctions (Historical Data)
Primary Dealer (own house account)	0.390	0.350	0.420	0.070
Direct Bidder (own house account)	0.040	0.030	0.050	0.020
Indirect Bidder (including foreign & intl monetary authorities)	0.540	0.510	0.590	0.080
FIMA Noncompetitive (foreign & intl monetary authorities)	0.000	0.000	0.000	0.000
Treasury Retail	0.010	0.010	0.010	0.000
Other Noncompetitive	0.000	0.000	0.000	0.000
SOMA (managed by Federal Reserve)	0.000	0.000	0.010	0.010

Insight: Primary Dealer and Indirect Bidder categories have the highest median values (0.39 and 0.54, respectively), indicating higher central tendency values compared to other categories. FIMA Noncompetitive, Other Noncompetitive, and SOMA have median values of 0, suggesting minimal or no activity in these categories. The Interquartile Range (IQR) is highest for Indirect Bidder (0.08), indicating more variability within this category compared to others. Treasury Retail has a consistent median, Q1, and Q3 value of 0.01, showing very little spread in the data.

Relative Comparison - Z score

The Z score is a metric to show how the allocations by bidder in the latest auction compares to historical norms. In so doing it provides insight into the relative performance of each investor type.

For this analysis the data is filtered to get the latest auction result. Then the mean and standard deviation of the allocation for each bidder type is computed. The z scores are then derived using a formula where the latest allocation for each bidder is subtracted from its mean allocation and divided by that bidder’s standard deviation.

Investor	Z-score
Z Scores by Investor
Latest Auction Allocation Z Scores
Primary Dealer (own house account)	0.647
Direct Bidder (own house account)	−0.238
Indirect Bidder (including foreign & intl monetary authorities)	−0.323
FIMA Noncompetitive (foreign & intl monetary authorities)	−0.397
Treasury Retail	−0.416
Other Noncompetitive	−0.833
SOMA (managed by Federal Reserve)	−0.710

Insight: In the latest auction Primary Dealers had a higher-than-average allocation. Direct Bidder, Indirect Bidder, FIMA Noncompetitive, Treasury Retail, Other Noncompetitive, and SOMA all had lower-than-average allocations, with Other Noncompetitive and SOMA showing the most significant deviations from the mean.

Primary Dealer (own house account): Z-score: 0.647 indicates that the allocation for Primary Dealers in the latest auction is 0.647 standard deviations above the mean. That suggests that Primary Dealers had a higher-than-average allocation in the latest auction.

Direct Bidder (own house account): Z-score: -0.238 indicates that the allocation for Direct Bidders in the latest auction is 0.238 standard deviations below the mean. This suggests that Direct Bidders had a slightly lower-than-average allocation in the latest auction.

Indirect Bidder (including foreign & intl monetary authorities): Z-score: -0.323 indicates that the allocation for Indirect Bidders in the latest auction is 0.323 standard deviations below the mean. This suggests that Indirect Bidders had a slightly lower-than-average allocation in the latest auction.

The remaining bidders all showed lower than average allocations.

Relative Comparison - Percentiles

Percentile ranks convey how each investor’s allocation in the latest auction compares to their historical allocations.

Investor	Percentile Rank
Allocation Percentile Ranks by Investor
Latest Auction Percentile Comparison
Primary Dealer (own house account)	76.9
Direct Bidder (own house account)	49.2
Indirect Bidder (including foreign & intl monetary authorities)	36.9
FIMA Noncompetitive (foreign & intl monetary authorities)	62.3
SOMA (managed by Federal Reserve)	31.5
Treasury Retail	34.6
Other Noncompetitive	16.2

Insight: The results show that Primary Dealers received a much stronger allocation relative to history that may indicate that they are being prioritized or that they submitted more competitive bids. Direct Bidders are in a middle ground, receiving what is essentially the historical norm. Indirect Bidders, SOMA, Treasury Retail, and Other Noncompetitive groups are receiving allocations that are on the lower end of their historical ranges.

Primary Dealer (own house account) – 76.9%: The latest allocation is higher than about 76.9% of all past allocations. This indicates that in the current auction, Primary Dealers received relatively more of the total allocation compared to what they usually receive historically. Direct Bidder (own house account) – 49.2%: Direct Bidder allocations in the latest auction are very close to the historical median. This means that the allocation is pretty much average when compared with previous auctions. Indirect Bidder (including foreign & intl monetary authorities) – 36.9%: The allocation for Indirect Bidders is on the lower side relative to their past performance. In roughly 63% of previous auctions, these bidders have received higher allocations.

FIMA Noncompetitive (foreign & intl monetary authorities) – 62.3%: Here, the allocation is above average. Around 62.3% of historical allocations were at or below the current allocation, suggesting that these bidders are getting a relatively higher share this time compared to many past auctions. SOMA (managed by Federal Reserve) – 31.5%: SOMA’s allocation in the latest auction is relatively low, as it ranks at the 31.5th percentile. This means that nearly 68.5% of historical auctions have seen SOMA receive a higher share. Treasury Retail – 34.6%: Similar to SOMA, Treasury Retail is at a lower percentile (34.6%), indicating that historically, they have received larger allocations more often than what was allocated in the latest auction. Other Noncompetitive – 16.2%: This is the lowest among all categories. An allocation at the 16.2nd percentile means that the vast majority (around 84%) of previous auctions have seen Other Noncompetitive bids receiving a larger allocation.

Trend Analysis

The code computes a 10 auction moving average for each investor. The latest auction result is then compared to that moving average to identify how the result compared to it. We see in the latest auction the marked deviation above the moving average for primary dealers.

Investor	Allocation	10-Auction Moving Avg	Deviation
Latest Auction Comparison
Allocation vs. 10-Auction Moving Average
Primary Dealer	0.426	0.354	0.072
Direct Bidder	0.040	0.043	−0.003
Indirect Bidder	0.521	0.590	−0.069
FIMA	0.000	0.000	0.000
Treasury Retail	0.007	0.006	0.000
Other Noncompetitive	0.003	0.003	0.000
SOMA	0.003	0.003	0.000

Insight: The primary dealer regression line (green) is sloping downward implying they have been receiving a smaller share of auction allocation over time in previous auctions. But the latest auction result (blue line) was well above the moving average showing a relative greater allocation. The gap between the blue line and moving average (red) showed a significant change.

The indirect bidder facet shows an upward sloping trend implying greater participation by that investor group. However the latest auction result showed a marked deviation from that trend and the moving average. Perhaps indicating weaker foreign and international demand.

Portfolio similarity

In this analysis each auction is considered a “portfolio” or a vector of allocations across all investor types (that add to 100%). We then compare the latest auction allocation vector to the historical allocation vector and each past auction vector.

The cosine similarity and Euclidean distance metrics are calculated to examine the relationship between latest auction allocation and previous allocations. Cosine similarity is a metric that ranges from -1 to 1 and indicates how similar the direction of the latest allocation vector is to the historical average (with 1 indicating identical orientation). Euclidean distance metric provides a direct distance measure between vectors with a lower value indicating more similarity.

Metric	Value
Similarity Metrics
Latest Auction vs. Historical Average
Cosine similarity (latest vs. historical average)	0.998
Euclidean distance (latest vs. historical average)	0.046

Issue Date	Cosine Similarity	Euclidean Distance
Similarity Metrics (Latest Auctions)
Cosine Similarity and Euclidean Distance for Recent Auctions
2025-03-04	0.991	0.090
2025-03-11	0.945	0.248
2025-03-18	0.953	0.224
2025-03-25	0.993	0.085
2025-04-01	0.990	0.099
2025-04-08	0.985	0.118

Insight: The two metrics (cosine similarity of 0.998 and Euclidean distance of 0.046) indicate that the allocation proportions are similar. Overall the latest auction allocation appears consistent with past auctions since the cosine similarity is close to 1 and Euclidean distance is low.

Graphical Representations

Box plot

A box plot is created for each investor showing the distribution of their allocations across all auctions of the 17 week bill. We mark the latest auction allocation as a distinct point.

Insight: The box plot highlights that the share of allocation for primary dealers is at the upper end of the interquartile range. While the indirect bidder allocation was lower than average.

Bar plot with error bars

The chart aims to highlight how the latest allocation deviates from the average while showing variability. The mean allocation for each investor is shown for all auctions with error bars showing 1 standard deviation. The latest auction allocation is identified as a diamond in red.

Insight: Once again it is shown that for primary dealers the allocation was above average but within the error bar. In the case of direct and indirect investors the allocation was within the average of past auctions.

Time series plot

Time series analysis plots the allocation for each investor over all auctions as separate lines with the latest auction marked with a vertical line. The chart reveals trends in the shift of allocation between investors over time.

Insight: Primary and indirect investors dominate the 17 week bill. The allocations have been broadly stable in recent auctions. Even though there was a spike up in primary dealer allocation in the latest auction it does not appear to be excessively beyond recent trends.

Radar/Spider chart

The plot shows the allocation profile across the seven investor groups with the latest auction result distinguished from the historical median allocations.

Insight: The chart highlights the sizable share of allocation between primary and indirect bidders followed by direct bidders. The relative less significance of non competitive investors and SOMA is clear.

Heatmap

In this analysis rows are auction count for the period that starts from 2018. Separate columns are shown for the different investor types and cell colors represent allocation sizes.

Insight: the chart highlights the dominant share by primary and indirect bidders once more. The red bar flags the latest auction’s allocation by investor. The stable color intensity for indirect bidders and primary dealers suggest stability over time in allocation.

Conclusion on buyer type: Based on the analysis of past auction allocation among buyer type the evidence is not conclusive that there was much deviation from past patterns. The primary bidder allocation went up but lay still within one standard deviation. The indirect bidder and direct bidder shares remain squarely within the average of past auctions. So on this measure of allocation shift by investor the result is inconclusive.

Notes

[^1] For the period between 2018-2025

References

U.S. Department of the Treasury. (2025). Announcements, results, and press releases. TreasuryDirect. https://www.treasurydirect.gov/auctions/announcements-data-results/announcement-results-press-releases/