Scenario wins: cassi (72) pgodzinbot (35) Mantic (27) smingers-bot (24) preseen (13) lewinke-thinking-bot* (11)
| Figure/Metric | Value | Source | Significance |
|---|---|---|---|
| 2026-27 Short Gilt Target | £95 Billion | DMO April 2026 Revision | Annual planned auction volume for 0-7 year gilts. |
| 2026-27 Medium Gilt Target | £56 Billion | DMO April 2026 Revision | Annual planned auction volume for 7-15 year gilts. |
| June 11 Short Auction Size | £5.0 Billion (Nominal) | DMO Announcement (June 4) | Confirmed size of the first Short auction in the window. |
| June 16 Medium Auction Size | £4.25 Billion (Nominal) | DMO Announcement (June 9) | Confirmed size of the first Medium auction in the window. |
| UK 10-Year Gilt Yield | 4.877% | Macrobond (June 5, 2026) | Current benchmark yield affecting auction pricing and cash raised. |
| Total Scheduled Auctions | 8 (5 Short, 3 Medium) | DMO Q1/Q2 Calendar | The universe of auctions used for the calculation. |
The UK’s debt issuance strategy has undergone a significant transformation following the September 2022 ‘Mini-Budget’ and the subsequent LDI crisis. Historically, the UK maintained a very long weighted-average maturity (WAM) compared to international peers (roughly 14-15 years versus 6-7 years for the US), supported by structural demand from pension funds. After the crisis, the DMO began shifting toward the short and medium ends of the curve to reduce volatility and accommodate changed demand. The 2026-27 Financing Remit continues this trend, with Long-dated conventional auctions effectively paused in favor of syndications. This historical pivot ensures that almost 100% of ‘Outright Auctions’ are now split between the Short (<7 years) and Medium (7-15 years) buckets, making the competition between these two sectors the primary determinant of the target variable.
My analysis for the share of “Medium” maturity gilts is primarily driven by the concrete auction schedule published by the UK Debt Management Office (DMO) and the strategic targets outlined in the 2026-27 Financing Remit. The DMO has signaled a structural shift away from long-dated debt following the 2022 Liability-Driven Investment (LDI) crisis, focusing instead on Short (0-7 years) and Medium (7-15 years) maturities. For the 2026-27 fiscal year, the DMO has set a target of £95 billion for short-dated auctions and £56 billion for medium-dated auctions. This suggests a full-year conventional auction ratio of approximately 37.1% for the Medium bucket.
However, the specific window from June 3, 2026, to August 11, 2026, contains a discrete set of eight scheduled conventional auctions. Based on the DMO’s quarterly calendar and recent announcements:
To estimate the cash-raised share, I analyzed the nominal sizes of these operations. The DMO has announced £5.0 billion for the June 11 short auction and £4.25 billion for the June 16 medium auction. Historically, short-dated conventional auctions tend to be sized slightly larger (£4.5–5.5 billion) than medium-dated auctions (£3.25–4.5 billion). Estimating total short-dated cash raised at approximately £23.0 billion and medium-dated cash raised at approximately £12.0 billion results in a central share of 34.3%.
Minor variations in this share will occur due to:
Question: what share of D3A “Total Cash Raised” in the 3 Jun–11 Aug 2026 window will come from Medium (vs Short/Long) conventional auctions? Forecasters agree the DMO calendar is the decisive input: inside the window there are five Short conventional auctions and three Medium ones (no Longs), and the 7 Jul 2033 gilt is Short under the question’s residual‑maturity formula (a common potential classification error). Using the 2026–27 remit and remaining‑auction averages implies a central medium share around the low‑to‑mid 30s percent (roughly 33–35%), consistent with announced nominal sizes so far (£5.0bn short on 11 Jun vs £4.25bn medium on 16 Jun). Forecasters warn that PAOF exercise (not available on the 2 Jul Green Medium gilt), auction prices/accrued interest (cash vs nominal), and any calendar revisions or unusually uneven sizes can move the final cash share by several percentage points; historical windows vary but are less directly comparable because they included Long auctions. Consensus places most probability mass in the low‑to‑mid 30s, with a modest downside tail into the high‑20s and upside toward ~40%; outcomes outside ~25–75% are considered very unlikely.
Forecast: Probability distribution:
Key driver is the mix of conventional gilt auctions (index-linked excluded) in the window 3 Jun 2026–11 Aug 2026 and their relative cash sizes. Three forecasters (1/3/4) independently identify the same eight in-window conventional auctions from the published DMO calendars: 11 Jun (2029), 16 Jun (2036), 24 Jun (2031), 2 Jul (Green 2037), 7 Jul (2033), 16 Jul (2036), 21 Jul (2029), 4 Aug (new Mar-2032). They also agree on the crucial bucket edge case: the 7 Jul 2026 auction of the gilt maturing 7 Mar 2033 has residual maturity 2033−2026+(3−7)/12 = 6.67 years, so it is Short (<7), not Medium. That leaves 3 Medium auctions (two 2036s and the Green 2037) and 5 Short auctions, with no Long auctions expected.
Given that composition, the Medium share is mainly determined by (a) the typical size differential (short auctions tend to be larger than medium auctions in the remit and recent practice), (b) any PAOF uplift (which would tend to benefit eligible conventional gilts; Green gilts are typically ineligible, slightly limiting Medium upside), and (c) low-probability calendar changes/cancellations or resizing. Using the announced sizes already known for early operations plus typical recent sizing for the remaining ones, a central Medium share in the low-to-mid 30s is most consistent with the evidence. I assign modest tail risk downward (e.g., unusually large short sizes / PAOF, or disruption to a medium auction) and upward (e.g., one or more short auctions moved out of window or materially upsized medium reopenings), but I view outcomes above ~45% as requiring multiple coincident surprises.
Forecaster 2’s distribution is materially higher (median ~44%) and appears to rely on a different/incorrect in-window auction mix (e.g., counting operations outside the window or assuming more medium weight than the calendars imply), so I down-weight it relative to the calendar- and maturity-rule-grounded analyses from forecasters 3 and 4, while keeping wider uncertainty than their tightest ranges to reflect execution and scheduling risk.
An amateur approach would often (i) use an annual remit share without checking the exact in-window auction set, (ii) split by auction count (3/8 = 37.5%) without accounting for systematically larger short-auction sizes, and/or (iii) misclassify the Mar-2033 gilt as “7 years” (Medium) by ignoring the month adjustment in the resolution formula. This forecast explicitly follows the resolution criteria (month-adjusted residual maturity; auctions only; index-linked excluded) and then models uncertainty from sizing/PAOF and potential schedule tweaks, producing a better-calibrated median near the low-to-mid 30s with realistic tails. Confidence in improvement is moderate-high because the main structural driver (3 medium vs 5 short auctions, with shorts typically larger) is robust to small pricing differences.
Forecast rationale (numeric):
— Iteration 1 — The forecasts share a common structure: they start from the DMO’s annual auction/remit mix, then adjust for the fact that this is a short, discrete 10-week window where a few auctions can materially change the share.
Overall, the forecasting logic points to a mid-30s to high-30s share, with the exact result depending on whether the 3 June–11 August calendar ends up being relatively short-heavy or whether the effective absence of long issuance boosts medium’s proportion. The dominant uncertainty is the lumpy, auction-by-auction nature of DMO issuance over a short window.
— Iteration 2 — Across the forecasts, the main reasoning pattern is that long-conventional gilt auctions are expected to be effectively absent or very limited over the window, so the relevant issuance mix is mostly Short vs. Medium buckets. On that basis, the Medium bucket is expected to account for less than half of total DMO auction issuance, with central estimates clustering in the low-to-mid 40% range.
There is strong agreement that the Medium share should be below 50%, with a central tendency around 40–45%.
The forecasts differ mainly on how short-tilted the schedule will be and whether any long-end auction unexpectedly reappears. That creates a fairly wide uncertainty band, even though the central view is fairly consistent.
— Iteration 3 — Across the forecasts, the core approach is to start from the DMO’s annual auction remit as a baseline and then adjust for how this specific 3 June–11 August window is likely to differ from the year-wide mix.
The collective reasoning centers on a baseline Medium share in the low-to-mid 30s, adjusted upward or downward depending on the auction calendar, with the most likely outcome viewed as somewhere between the low 30s and around 40%, and the main uncertainty coming from the mix of Short vs. Medium auctions in a small number of scheduled sales.
Scheduled Auctions and Maturity Classification Forecasters universally agree that based on the UK Debt Management Office’s (DMO) official operational calendar, there are exactly eight outright conventional gilt auctions scheduled within the target window. Index-linked gilts, programmatic tenders, and syndicated offerings are consistently excluded from this count per the resolution criteria. Applying the strict residual maturity formula provided in the prompt, forecasters classify the eight operations into two specific buckets:
Estimating Total Cash Raised The resolution depends on the actual “Total Cash Raised” rather than nominal issuance sizes or a simple count of the auctions. Forecasters highlight two structural factors that will determine the final cash volume:
The Post-Auction Option Facility (PAOF) and Uncertainties A central variable in projecting the final cash raised is the PAOF, which allows primary dealers to purchase additional gilt allocations post-auction at the clearing price.
Overall, the rigid operational schedule (five short versus three medium auctions) combined with the DMO’s structurally larger sizes for short-end operations forms a highly predictable baseline. Variance is driven almost entirely by minor pre-auction sizing adjustments and dynamic post-auction market demand.
Question: What share of UK government debt issued via DMO auction between 3 June 2026 and 11 August 2026 will fall in the “Medium” maturity bucket (7–15 years)? Final Prediction: Probability distribution:
Total Cost: extra_metadata_in_explanation is disabled Time Spent: extra_metadata_in_explanation is disabled LLMs: extra_metadata_in_explanation is disabled Bot Name: extra_metadata_in_explanation is disabled
Forecaster 1: Probability distribution:
Forecaster 2: Probability distribution:
The research finds that the primary anchor for forecasting the share of conventional gilt auction proceeds in the Medium (7–15 years) bucket is the DMO’s revised April 23, 2026 financing remit, which allocates £56.0bn of £158.4bn of conventional gilt auctions to Medium maturities (35.4%). It documents a structural shift since 2022 toward shorter issuance (Short <7 yrs: 60.0%; Long ≥15 yrs: 4.7%), notes recent auction evidence around the window (examples: May 21 2036 gilt — Medium; May 28 2030 gilt — Short; June 11 2029 gilt — Short; June 16 2036 gilt — Medium), and reports the DMO’s July–September plan for 13 regular gilt auctions (implying ~8–12 conventional auctions likely fall in the June 3–Aug 11 window). The research lists factors that could push the Medium share above or below the 35.4% anchor (e.g., recent sizeable Medium auctions vs. strong demand for Short gilts and cost/refinancing considerations), and notes limitations: no specific auction calendar for June 3–Aug 11, 2026; no prediction-market forecasts found; and a lack of month-by-month historical maturity-mix patterns.
Sources used (as stated in the research): https://www.dmo.gov.uk https://www.bankofengland.co.uk https://www.niesr.ac.uk https://manifold.markets https://www.metaculus.com https://polymarket.com
You’re forecasting what percentage of UK conventional gilt auction proceeds between June 3-August 11, 2026 will be in the Medium maturity bucket (7-15 years residual maturity at issuance).
The most relevant base rate comes from the revised April 23, 2026 financing remit [12]:
Conventional Gilt Auctions for Full Year 2026-27:
This 35.4% medium share for the full fiscal year is your primary outside-view anchor.
The DMO has undergone a dramatic strategic pivot since the September 2022 LDI crisis [13][15]:
Identified auctions near your time window:
This recent pattern shows alternating short/medium auctions with substantial sizes for both buckets.
Per DMO announcement [3]:
This suggests roughly 4-5 conventional auctions per month during summer, meaning your 10-week window (June 3-August 11) likely captures approximately 8-12 conventional auctions.
[1][15] note that:
Primary anchor: 35.4% medium share (from annual remit) [12]
Historical context:
Seasonal considerations:
Reasons Medium share might be HIGHER than 35.4%:
Reasons Medium share might be LOWER than 35.4%:
What I could NOT find:
Data reliability:
Your central tendency should anchor around 35.4% (the annual remit allocation), but with substantial uncertainty given:
The lack of prediction markets or crowd forecasts means you’re working from first principles and official data without market-based aggregation to cross-reference.
Conditional on the question not being annulled.
The formal window closes at 11 August 2026 23:59 UTC, about 63 days from today, 9 June 2026. On the current DMO calendar, the last included conventional auction is 4 August 2026, so the practical answer should be knowable roughly 56 days from now unless the calendar changes. (dmo.gov.uk)
The best base rate is the 2026-27 revised remit auction mix: £95.0bn short, £56.0bn medium, and £7.4bn long conventional gilt auctions. That implies a medium share of 56.0 / (95.0 + 56.0 + 7.4) = 35.4% for conventional gilt auctions across the fiscal year. (dmo.gov.uk)
The specific calendar for this resolution window points a bit below the 35.4% annual auction base rate. From 3 June through 11 August, the included conventional auctions currently appear to be 5 short auctions and 3 medium auctions, with no long conventional auction in the window. The relevant scheduled medium auctions are 16 June 2036, 2 July Green Gilt 2037, and 16 July 2036; the relevant scheduled short auctions are 11 June 2029, 24 June 2031, 7 July 2033, 21 July 2029, and 4 August 2032. (dmo.gov.uk)
If the calendar and usual auction sizing remain broadly unchanged, I expect a medium share around 33%. This is below the annual base rate mainly because the window cuts off before the 18 August 2036 medium auction and includes several large short auctions; also the 2 July Green Gilt 2037 auction is medium but has no PAOF, which likely modestly lowers medium cash raised relative to comparable non-green auctions. (dmo.gov.uk)
Continuing the recent short/medium split and recent auction size pattern gives roughly 32% to 34%. The DMO’s recent outright auction report shows many conventional auctions around £4.0bn to £5.0bn nominal, recent medium 2036 and green 2037 auctions in the same neighborhood but not clearly larger than shorts, and recent total cash raised close to nominal after price and PAOF effects. (dmo.gov.uk)
I do not see a useful prediction-market price for this exact quantity, so the main “expert/market” signal is the DMO calendar produced after consultations with GEMMs and investors. The consultation/final calendar process shows continued support for both short and medium issuance, but the final Q2 calendar removed the proposed July long conventional auction from the auction schedule and scheduled a mix that still leans short by count. (dmo.gov.uk)
A low outcome could occur if one or more medium auctions are reduced, postponed, or shifted into tenders/syndications, while the short auctions—especially the 2029 and new 2032 issues—are sized aggressively and have strong PAOF take-up. That could push the medium share into the low-to-mid 20s.
A high outcome could occur if the DMO sizes the 2036 reopenings and/or Green Gilt 2037 materially larger than expected, while short auctions are smaller or see little PAOF take-up. A more extreme version would be cancellation/postponement of a short auction but not medium auctions, pushing the share toward or above 40%.
I anchor on the 35.4% annual conventional-auction medium share from the revised remit. The inside view moves me modestly downward because the known window is not a representative slice of the full year: it currently has 3 medium vs. 5 short included conventional auctions, and one of the three medium auctions is a green gilt, where PAOF is not available. The 11 June 2029 short auction is already announced at £5.0bn nominal, and the 16 June 2036 medium auction has been announced at £4.25bn nominal, consistent with short auctions being at least as large as medium auctions on average. (dmo.gov.uk)
My central estimate is therefore about 33% rather than 35.4%. Most normal variation comes from unknown sizes for the 24 June, July, and 4 August auctions; realized prices; and PAOF usage. I put most mass between 30% and 37%, with tails for calendar alteration, cancellation, or unusual sizing.
Percentile 0.1: 16.0
Percentile 1: 22.5
Percentile 5: 27.8
Percentile 10: 29.7
Percentile 20: 31.2
Percentile 30: 32.0
Percentile 40: 32.6
Percentile 50: 33.0
Percentile 60: 33.7
Percentile 70: 34.5
Percentile 80: 35.5
Percentile 90: 37.2
Percentile 95: 39.3
Percentile 99: 46.5
Percentile 99.9: 60.0
(a) Time left until known: Conditional on the question not being annulled or ruled ambiguous, the last currently scheduled included auction is on 4 August 2026, so the outcome should be effectively known in about 8 weeks from 9 June 2026; the formal window closes on 11 August 2026, about 9 weeks away. The 12 August index-linked auction is outside the window and excluded. (dmo.gov.uk)
(b) Outside view: The strongest base rate is the 2026-27 auction remit: £95.0bn short, £56.0bn medium, £7.4bn long conventional auction sales, implying a medium share of 56.0 / (95.0 + 56.0 + 7.4) = 35.4% of conventional auction cash. (dmo.gov.uk) The latest current-remit table as of 9 June 2026 keeps the same planned auction amounts, with medium auction proceeds to date of £7.108bn and short auction proceeds to date of £18.012bn including PAOF. (dmo.gov.uk)
(c) Inside view: The published calendar is now quite informative. Within the window, the scheduled conventional auctions appear to be 5 short auctions and 3 medium auctions: short on 11 Jun, 24 Jun, 7 Jul, 21 Jul, 4 Aug; medium on 16 Jun, 2 Jul, 16 Jul. (dmo.gov.uk) That count alone might suggest around 37.5% medium, but medium cash sizes are likely a bit smaller because one medium auction is 4⅝% Green Gilt 2037, and green gilt auctions have no PAOF, while several short auctions are likely benchmark-building operations near £4bn-£5bn. (dmo.gov.uk) The 11 June short auction is already announced at £5.0bn nominal of 4% Treasury Gilt 2029, which pushes me slightly below the 35.4% base rate. (dmo.gov.uk)
(d) Outcome if nothing changed: If the DMO follows the published calendar and future auctions average roughly the remaining-remit cash need per auction, I get about 34% medium.
(e) Outcome if the current trend continued: Recent 2026 auctions show high short issuance, meaningful but not dominant medium issuance, and little/no long conventional auction issuance in this period; carrying that forward gives roughly 32%–35% medium. (dmo.gov.uk)
(f) Expectations of experts and markets: I found no prediction-market price for this exact question. The closest “expert” evidence is the DMO’s remit and consultation record: GEMMs and investors supported multiple short reopenings and at least two 2036 medium reopenings, plus a Green Gilt 2037 reopening, which matches the published short/medium-heavy calendar. (dmo.gov.uk)
(g) Unexpected low scenario: Short auctions are upsized or get heavy PAOF take-up, while the 2036 medium reopenings are kept near £4bn with little/no PAOF and the Green Gilt 2037 auction is small. That can push the share into the low-to-mid 20s.
(h) Unexpected high scenario: One short auction is reduced, delayed, or fails to attract PAOF, while both 2036 medium auctions are upsized and receive strong PAOF take-up. A calendar change adding another medium auction before 11 August would be the clearest path above 45%.
Rationale: I anchor on the 35.4% auction-remit base rate. The key update down is the specific window composition: 5 short versus 3 medium auctions, with a £5bn short auction already announced and one of the medium auctions being a green gilt with no PAOF. The key update up is that long auctions are essentially absent from the window, so the denominator is mostly short and medium rather than being diluted by long. Netting those, I put the median slightly below the remit base rate, around 33%–34%, with most uncertainty coming from auction sizes, prices, and PAOF take-up rather than from which gilts will be auctioned.
Percentile 0.1: 20.0
Percentile 1: 23.5
Percentile 5: 26.8
Percentile 10: 28.5
Percentile 20: 30.5
Percentile 30: 31.8
Percentile 40: 32.8
Percentile 50: 33.7
Percentile 60: 34.6
Percentile 70: 35.7
Percentile 80: 37.0
Percentile 90: 39.5
Percentile 95: 42.0
Percentile 99: 48.0
Percentile 99.9: 57.0
Modal: 31.3-37.5 (61.0%) • partial_shift • 6m9s
Reading: strict
The resolution criteria are highly specific and mechanical: exact date boundaries, a precise maturity formula, named DMO report codes (D3A, D1A), explicit bucket boundaries (7 and 15 years), and a defined rounding rule (one decimal place). There is little room for charitable interpretation — the resolver must follow the stated formula exactly, leaving almost no discretion.
Total evidence registered (shared pool): ?
| Variant | Perspective | Model | Turns | Tools | Status |
|---|---|---|---|---|---|
| 0 | inside_view (inside_view_v1) | openai/gpt-5-mini | 25 | 25 | OK |
| 1 | outside_view (outside_view_v1) | anthropic/claude-sonnet-4-6 | 19 | 30 | OK |
| 2 | contrarian (contrarian_v1) | openai/gpt-5 | 11 | 11 | OK |
Evidence confidence: high
high evidence]All eight confirmed in-window conventional auctions proceed as planned. The three Medium auctions (16-Jun TG2036, 2-Jul Green2037, 16-Jul TG2036) raise amounts close to the remit average (~£4.25bn each). The five Short auctions raise amounts near their respective planned sizes (~£4.75–5.0bn each). No additional medium auction is inserted and no planned auction is cancelled. The 7-Jul TG2033 remains in the Short bucket (residual 6.67 years < 7 years).
If the DMO adds an unscheduled medium auction, inserts a long-conventional auction, or adjusts sizes materially toward medium, the share could rise above this range. If short auction sizes are significantly smaller than assumed, the medium share rises.
low evidence]DMO inserts an additional medium-maturity auction date not in the current Q2 calendar (e.g. a reopening of a medium gilt or a new medium-dated issuance). Alternatively, the sizes of the three confirmed medium auctions are set materially higher than the remit average (e.g. £5.5–6bn each), while short sizes remain near plan.
Current Q2 calendar shows no unscheduled medium slots; adding auctions requires a new DMO press release. Full-year remit caps total medium issuance at £56bn across 15 auctions, leaving limited room for size inflation. The in-window window covers roughly 2.5 months (~20% of fiscal year), and the remit pacing is roughly linear.
medium evidence]The 4-Aug new short gilt (TG2032, ~5.58 years) and other short auctions are sized larger than plan (e.g. £5.5–6bn each). The DMO inserts another short-maturity auction. One of the three medium auctions is cancelled or has its size cut, reducing medium cash raised.
Current confirmed data shows three medium auctions scheduled at typical sizes; no indication of cancellations. OBR analysis (source 11) shows structural shift toward short but medium is still a significant and planned proportion. Remit ratio of 35% medium provides a gravitational anchor.
inside_view_v1)Inside-view causal chain analysis:
The evidence quality is rated “high” by the brief, and the auction calendar for the window is essentially known at this stage (2026-06-09).
Confirmed auction structure (authoritative DMO sources):
- 8 conventional gilt auctions in the 3 Jun–11 Aug window
- 3 Medium bucket: 16-Jun (4⅞% TG2036, ~9.6 yr), 2-Jul (4⅝% Green2037, ~10.67 yr), 16-Jul (4⅞% TG2036 reopening, ~9.5 yr)
- 5 Short bucket: 11-Jun (TG2029, ~2.92 yr), 24-Jun (TG2031, ~4.75 yr), 7-Jul (TG2033, ~6.67 yr — SHORT, confirmed by sources 14–16), 21-Jul (TG2029, ~2.83 yr), 4-Aug (new TG2032, ~5.58 yr)
- 0 Long bucket (no long conventional auctions in window)
- 23-Jul removed as outright auction (programmatic tender instead, confirmed authoritative sources 17–18)
Sizing estimates:
- 16-Jun auction: £4,250mn confirmed (source 8, ainvest citing DMO)
- 11-Jun auction: £5,000mn confirmed (source 7)
- Remit average: ~£3.73bn/medium auction, ~£4.75bn/short auction
- If medium auctions average ~£4.25bn each: 3 × £4.25bn = £12.75bn medium
- If short auctions average ~£4.75–5.0bn each: 5 × £4.875bn = £24.375bn short
- Total ≈ £37.1bn; Medium share ≈ 12.75 / 37.1 ≈ 34.4%
High-evidence scenario (relative_evidence_strength: high): Medium share 33–37%. The three medium auctions at ~£4.25bn each and five short auctions at ~£4.75–5bn each lands squarely in the 31.3–43.8% range, with the central mass around 33–37%.
Medium-evidence scenario (relative_evidence_strength: medium): Below 30% if short auctions are larger or additional short dates added. Some possibility but no indication in current calendar.
Low-evidence scenario (relative_evidence_strength: low): Above 40% if additional medium auctions inserted. No current signal.
Key causal mechanism: Current state = 3 medium / 5 short auction schedule confirmed → at remit-consistent sizing → outcome ≈ 34% medium share.
The 7-Jul TG2033 (6.67 years) is firmly Short (sources 14-16 authoritative), so this won’t shift to Medium. The remit anchors the sizing ratio. The central estimate is ~34-35%, well inside the 31.3–37.5% (bin_2) range.
Distribution construction:
- 31.3–37.5% (bin_2): highest mass, ~45% — the high-evidence scenario lands here
- 37.5–43.8% (bin_3): ~20% — if medium auctions are sized above remit average
- 25–31.3% (bin_1): ~18% — if short auctions are sized larger or extra short auction added
- Below 25% (bin_0): ~5% — tail risk of two medium auctions cancelled/very small
- 43.8–50% (bin_4): ~7% — additional medium auction or significantly higher medium sizing
- 50–56.3% (bin_5): ~2%
- bin_6+: ~3% total for very high outcomes
- The low-evidence “above 40%” scenario gets materially less mass than the high-evidence central scenario.
outside_view_v1)Reference class: UK DMO conventional-auction maturity composition for the 2026–27 Financing Remit period. Base rate: the revised 2026–27 remit plans £56.0bn medium vs £95.0bn short (≈35% medium by value) (brief source 2). Adjustment: only small adjustments from the base rate because the in-window auction slate closely matches remit pacing — the Q2 calendar lists three medium auctions (16-Jun TG2036, 2-Jul Green 2037, 16-Jul TG2036) and five short auctions (11-Jun TG2029, 24-Jun TG2031, 7-Jul TG2033 (residual ≈6.67y → Short per brief source 16), 21-Jul TG2029, 4-Aug new TG2032) (brief sources 5,13,16,17–18). The 16-Jun size is known (~£4.25bn) and 11-Jun was £5bn (brief sources 8,7), which supports remit-like sizing. Risks that push the share down (medium auctions cut or short auctions oversized) or up (an extra medium auction or oversized medium reopenings) are possible but not well-supported: the brief’s scenario set tags the three-medium / five-short outcome as high-evidence, an above‑40% medium outcome as low-evidence, and a below‑30% medium outcome as medium-evidence.
Final — distributional judgment (outside view, weighted by evidence strength): place most mass on the remit-consistent band (31.3–37.5%, bin_2) because the authoritative remit and Q2 calendar anchor the in-window composition (brief sources 2,5,17–18) — this scenario is high-evidence and therefore carries materially more probability. Allow a meaningful tail to the left (25–31.3%, bin_1) to reflect plausible sizing/cancellation risk (medium-evidence), a small tail to the right (37.5–43.8%, bin_3) to reflect the low-probability insertion/upsizing of medium auctions (low-evidence), and a very small probability for extreme outcomes below 25% (bin_0) given the explicit existence of three medium auctions in the window. This yields the forecasted probability mass: bin_2 dominant (0.74), bin_1 moderate (0.19), small mass in 0.06 (bin_3) and 0.01 (bin_0).
Caveats / strict-reading note: I follow the question’s strict mechanical rules (resolution reads DMO D3A and D1A exactly; the 7-Jul TG2033 residual ≈6.67 years is Short per those rules — brief source 16). That strict reading reduces ambiguity around borderline maturities and supports concentrating mass near the remit-derived base rate.
contrarian_v2)Adversarial reading
The implicit consensus in the brief is that the medium share will be around 33-37% given the 3:5 medium-to-short auction ratio and remit-consistent sizing. The bulk of the evidence (8 confirmed dates, specific gilt identifications, boundary confirmations like 6.67 years for TG2033) pulls strongly toward this.
Counter-evidence found (or absence)
Searched for counter-evidence; nothing substantive found; consensus appears robust. The low-evidence scenario for >40% relies on unannounced additions, which the Q2 calendar shows no room for without new releases. The medium-evidence <30% scenario would require deviations from planned sizes that contradict the remit pacing.
Distribution implications
The high-evidence scenario gets the bulk of the mass, concentrated in 31.3-37.5 (bin_2) and 37.5-43.8 (bin_3) with a slight tilt toward bin_2. Minimal mass on extremes because counter-evidence is thin. This aligns with the adversarial test surviving scrutiny.
Strategy: logit_mean • Drafts: 3/3
anchoring_v1The draft rationale correctly identifies the 3-medium / 5-short auction structure and derives a central estimate near 34%. The primary anchoring concern is that the draft places ~60% of probability mass in 31.3–37.5% (bin_2), which actually represents the modal expectation well — this is not obviously optimistic. However, the draft’s scenario construction subtly anchors toward the optimistic end of the sizing range by assuming medium auctions average ~£4.25bn (using the one known confirmed medium size from 16-Jun) while assuming short auctions average only ~£4.75–5.0bn, when the one confirmed short size (11-Jun at £5.0bn) and the remit average for shorts (£95bn/20 = £4.75bn) are both consistent with shorts being larger. Additionally, the brief notes the full-year remit short-to-medium ratio is ~60/35 by value, which would imply a medium share closer to 37% of the two-bucket total — but the five-short / three-medium count ratio alone implies ~37.5% only if sizing is equal, and short auctions have historically been sized larger than medium auctions in this remit cycle. With shorts averaging £5bn vs mediums at £4.25bn, the arithmetic central case is 3×4.25 / (3×4.25 + 5×5.0) = 12.75/37.75 = 33.8%, firmly in bin_2. The draft’s distribution is broadly appropriate, though the left-tail probability (bin_1: 25–31.3%, 16.4%) may be slightly light given that shorts being consistently oversized versus plan is the medium-evidence scenario flagged in the brief.
ceiling_v1The brief provides a near-complete structural picture of the window: exactly 8 confirmed conventional auction dates, with 3 unambiguously classified as Medium and 5 as Short. This is a hard structural constraint on the outcome. The key ceiling/floor arithmetic: if all auctions are sized at remit averages (£3.73bn medium, £4.75bn short), medium cash = 3 × £3.73bn = £11.19bn; short cash = 5 × £4.75bn = £23.75bn; total = £34.94bn; medium share = 32.0%. With the confirmed 16-Jun size of £4.25bn applied to all three medium auctions (£12.75bn medium) and short auctions at ~£4.875bn average (£24.375bn short), the share = 34.3%. To reach 37.5–43.8% (bin_3), medium auctions would need to average ~£5.3–6.2bn while short auctions stay near £4.75bn — materially above the remit average of £3.73bn for medium. The structural ceiling from the remit constraint (£56bn medium / 15 auctions = £3.73bn avg) makes values much above 37% structurally unlikely unless sizing is anomalously large for medium. The draft allocates 17.2% to 37.5–43.8% (bin_3) and 2.0% to 43.8–50% (bin_4), which appears slightly over-generous given the ceiling implied by remit average sizing. Mass in 31.3–37.5% (bin_2) is well-supported and correctly dominant, but the upper tail (bin_3 onward) should be trimmed modestly in favor of bin_2 given the confirmed 3-medium/5-short structural composition with remit-capped sizing.
math_v1The forecaster’s distribution is well-anchored to remit-derived reasoning (3 medium auctions at ~£4.25bn each, 5 short at ~£4.75–5.0bn each, implying ~34% medium share), but the actual probability allocation across bins exhibits two internal inconsistencies. First, the reported “high-evidence scenario” of 33–37% should map to 31.3–37.5 (bin_2) with dominant mass (~45% per rationale), yet bin_2 receives only 59.8%, leaving 40.2% scattered across other bins despite the high-evidence forecast centering there. Second, the distribution sums to 1.000 ✓ but the per-bin values appear to derive from a logit-transformed averaging of multiple internal models that has not been re-normalized to match the stated 33–37% central tendency; the tail mass (bin_0: 2.3%, bin_1: 16.4%) suggests hedging against downside scenarios that the brief rated only “medium-evidence,” contradicting the emphasis on the “high-evidence” central band. The shape is defensible but does not fully cohere with the explicit reasoning.
Reviews consumed: 3
All three reviewers broadly endorsed the draft’s central tendency near 33–35% (bin_2). The ceiling reviewer flagged that remit-average sizing (£3.73bn medium vs £4.75bn short) creates a structural ceiling making bin_3+ over-generous in the draft, and that the 3M/5S confirmed composition with typical sizing makes the bin_0 tail over-generous; I accepted both and shifted mass from 2.3%→1.5% (bin_0) and bin_3/4 into bin_2. The math reviewer flagged that bin_2 should carry more dominant mass given the high-evidence central scenario tag, and that lower-tail hedging (18.7%) was disproportionate to evidence — I partially accepted, concentrating bin_2 to 0.61. The anchoring reviewer suggested no_change but noted possible size-asymmetry risk that could push closer to ~30%; I kept meaningful bin_1 mass (0.15) to honor that. The result is a partial shift: more concentrated on bin_2, slightly trimmed tails.
[math] Lower-tail mass (bin_0+bin_1=18.7%) too high relative to brief’s evidence weighting; should concentrate even more in ~0.65 (bin_2).
[anchoring] Size-asymmetry: if shorts systematically run at £5bn upper bound while mediums run at remit average £3.73bn, share could fall to ~30%, requiring more left-tail mass.
| Bin | Probability |
|---|---|
| Below 25 (bin_0) | 1.5% |
| 25-31.3 (bin_1) | 15.0% |
| 31.3-37.5 (bin_2) | 61.0% |
| 37.5-43.8 (bin_3) | 15.0% |
| 43.8-50 (bin_4) | 4.0% |
| 50-56.3 (bin_5) | 1.5% |
| 56.3-62.5 (bin_6) | 0.8% |
| 62.5-68.8 (bin_7) | 0.5% |
| 68.8-75 (bin_8) | 0.4% |
| Above 75 (bin_9) | 0.3% |
Skip to main content
You can find information on the gilt market (including green gilt issuance), and find out more about buying and selling gilts.
Find out about the annual financing remit set by HM Treasury for the DMO, as well as key announcements and publications relating to the DMO’s remits since 1998.
Information on the DMO’s cash management activities, including money market instruments.
Find out about the government’s green gilt issuance.
PWLB’s function is t
wooley_rhino • run b4305346 • $2.86 • 6m9s
Extracted Percentile Values: {0.1: 24.0, 1.0: 26.5, 5.0: 29.5, 10.0: 31.5, 15.0: 33.0, 20.0: 34.2, 25.0: 35.0, 30.0: 35.8, 35.0: 36.6, 40.0: 37.3, 45.0: 37.8, 46.0: 37.9, 47.0: 38.1, 48.0: 38.3, 49.0: 38.5, 50.0: 38.7, 51.0: 38.9, 52.0: 39.1, 53.0: 39.3, 54.0: 39.5, 55.0: 39.7, 60.0: 40.7, 65.0: 42.0, 70.0: 43.5, 75.0: 45.5, 80.0: 48.0, 85.0: 51.0, 90.0: 55.0, 95.0: 60.5, 99.0: 74.0, 99.9: 76.5}
Model’s Answer: My forecast centers on the high 30s.
Step one is weighting the sources. The most important evidence is the official DMO remit and the official quarterly auction calendars. Those matter much more than commentary because this question is almost entirely about the DMO’s scheduled maturity mix, not about a market price that can move continuously. The Investing.com and press-notice items are useful mainly because they confirm specific June auctions and sizes.
Step two is establishing the base-rate mix. The 2026-27 remit allocates conventional auction issuance roughly as short 97.3 bn, medium 57.8 bn, long 8.0 bn. That implies a full-year medium share of about 35.5 percent of conventional auction cash if the fiscal-year mix were followed mechanically. If long conventional issuance is effectively absent in this window, then the relevant short-versus-medium baseline becomes about 37.3 percent.
Step three is adjusting for the actual window. The biggest detail is that the 2 June 2037 green gilt auction, which is medium bucket, sits just outside the window and is excluded. Inside the window, the known June auctions are short on 11 June, medium on 16 June, and short again on 24 June. So the opening part of the measurement period is already a bit short-heavy. That pushes the window-specific forecast somewhat below what a naive “medium is about 38 percent of short-plus-medium issuance” rule would otherwise suggest.
Step four is considering July through 11 August. The official context strongly suggests the DMO continues to issue mostly in short and medium buckets, with little or no long conventional supply. The key uncertainty is whether July and early August are closer to balanced between short and medium, or still short-skewed. If they are roughly balanced, the overall window ends up around the upper 30s or low 40s. If they are more short-heavy, it lands in the low-to-mid 30s. A very high outcome would need a run of medium auctions later in July and early August, which is possible but not my base case.
The assumptions most likely to matter are these. First, that there is no meaningful long conventional auction in the window; I put about a 20 percent chance on that being wrong. Second, that July and early August conventional auctions are roughly balanced or only modestly short-tilted rather than strongly short-heavy; I put about a 30 percent chance on that being wrong. Third, that cash-raised amounts stay near the usual size pattern, with short auctions a bit larger on average than medium; I put about a 20 percent chance on that being wrong. The second assumption is the biggest driver of forecast error.
A low outcome would come from several additional short re-openings in July and early August, with few medium auctions after the known 16 June sale. That could pull the result into the upper 20s or low 30s. I think below 25 percent is very unlikely, around 1 percent or less, but not impossible.
A high outcome would come from multiple additional 10-year-area auctions, perhaps including another medium green or medium conventional reopening, with only one more short auction before 11 August. That could lift the result into the 50s. Above 75 percent seems extremely unlikely, but the question requires some tail there.
Overall, the short horizon and the administrative nature of the outcome make me anchor fairly tightly on the known remit and calendar logic, while still leaving meaningful uncertainty because the exact July and early August maturity choices are not fully pinned down in the material provided.
Percentile .1: 24.0 Percentile 1: 26.5 Percentile 5: 29.5 Percentile 10: 31.5 Percentile 15: 33.0 Percentile 20: 34.2 Percentile 25: 35.0 Percentile 30: 35.8 Percentile 35: 36.6 Percentile 40: 37.3 Percentile 45: 37.8 Percentile 46: 37.9 Percentile 47: 38.1 Percentile 48: 38.3 Percentile 49: 38.5 Percentile 50: 38.7 Percentile 51: 38.9 Percentile 52: 39.1 Percentile 53: 39.3 Percentile 54: 39.5 Percentile 55: 39.7 Percentile 60: 40.7 Percentile 65: 42.0 Percentile 70: 43.5 Percentile 75: 45.5 Percentile 80: 48.0 Percentile 85: 51.0 Percentile 90: 55.0 Percentile 95: 60.5 Percentile 99: 74.0 Percentile 99.9: 76.5
UK gilt stock carries unusually long weighted-average maturity, historically supported by long-dated pension and insurance demand that has since weakened.
Official supply intent has shifted away from the long end, concentrating issuance into Short and Medium buckets over the window.
The auction calendar between 3 June and 11 August contains five Short conventional auctions and three Medium conventional auctions, producing a Medium share of about 33.1%.
Expected cash weights reflect announced sizes, modal green-gilt sizing and PAOF conventions, which reduce Medium cash relative to nominal counts.
A single mis-bucketed 2033 sale that crosses the 7‑year cutoff would raise Medium share by roughly 11 percentage points.
High PAOF exercise on eligible auctions, a larger-than-expected green-gilt or a late-added long conventional auction would each pivot the share materially.
Primary uncertainties are unannounced auction amounts, PAOF take-up and last‑minute calendar format changes such as auction versus tender.
Sensitivity concentrates on the 7‑year residual‑maturity boundary and PAOF treatment at green auctions; small timing or sizing moves produce nonlinear shifts in the percentage.
Forecast (median): 38.8652%
I start from the official auction calendars, because the resolution only counts conventional gilt auctions in the window 3 June 2026 through 11 August 2026. As of 9 June 2026, the relevant conventional auctions scheduled in-window are: 11 June (4% Treasury Gilt 2029), 16 June (4⅞% Treasury Gilt 2036), 24 June (4⅛% Treasury Gilt 2031), 2 July (4⅝% Green Gilt 2037), 7 July (4⅛% Treasury Gilt 2033), 16 July (4⅞% Treasury Gilt 2036), 21 July (4% Treasury Gilt 2029), and 4 August (a new conventional gilt maturing 7 March 2032). The 3 June and 12 August auctions are index-linked and excluded, and the 23 July operation was shifted from an auction to a programmatic tender. Using the maturity dates in the prospectuses/calendar, the medium-bucket auctions in this window are 16 June 2036, 2 July 2037, and 16 July 2036; the other five are short. (dmo.gov.uk)
That means the count-based starting point is simple: 3 of the 8 included auctions are medium, so if all auctions raised the same cash the answer would be 37.5%. But the remit is not size-neutral across sectors. In the revised 2026-27 remit, planned auction sales are £95.0 billion for short gilts and £56.0 billion for medium gilts. The DMO’s progress table as of 9 June 2026 shows £76.988 billion of short auction sales still required across 16 remaining short auctions versus £48.892 billion across 12 remaining medium auctions, i.e. about £4.81175 billion per remaining short auction and £4.07433 billion per remaining medium auction. Applying those remaining-sector averages mechanically to the 5 short and 3 medium auctions inside this question’s window gives an implied medium share of 33.6891%, already below the naive 37.5% count share. (dmo.gov.uk)
I then adjust a bit lower using the specific securities in this window. The DMO’s official outright-auction results report as of 9 June 2026 shows recent cash raised of £3.952 billion for the 21 May 2026 auction of 4⅞% Treasury Gilt 2036 and £3.156 billion for the 2 June 2026 auction of 4⅝% Green Gilt 2037. By contrast, recent short auctions raised £4.966 billion for 4% Treasury Gilt 2029 on 21 April, £4.719 billion for 4⅛% Treasury Gilt 2031 on 12 May, and £3.901 billion for 4⅛% Treasury Gilt 2033 on 27 May; short auctions also showed meaningful PAOF take-up while medium auctions in the current fiscal year did not. On top of that, the already-announced 11 June auction of 4% Treasury Gilt 2029 is sized at £5.0 billion nominal, which points to a relatively large short denominator early in the window. (dmo.gov.uk)
There are still upside risks to the medium share. The DMO says auction sizes are set case-by-case and usually announced only one week ahead, and the revised remit says the 2026-27 green gilt programme is intended to focus on re-opening and building 4⅝% Green Gilt 2037 to benchmark size. So a larger-than-expected 2 July green auction, or larger 16 June / 16 July 2036 re-openings, could push the share back into the mid-30s. The main downside risk is that short auctions continue to be sized more aggressively than medium ones, especially if the 4 August launch of the new 7 March 2032 gilt is large and if short PAOF uptake is healthy. (dmo.gov.uk)
My synthesis is a point estimate of 32.3%. That is below both the 37.5% count-based share and the 33.6891% remit-average share because the actual mix inside the window is tilted toward relatively large short auctions (especially 2029 and the new 2032) and relatively modest medium auctions (two 2036 re-openings plus one green 2037 reopening). I would think of the distribution as centered in the low 30s, with an 80% interval of roughly 29.5% to 35.5%, and only small tails below about 28% or above about 38.5%. (dmo.gov.uk)
To resolve this I first map the scheduled DMO conventional auctions in scope and bucket them using the resolution formula. The 2 June 2026 auction of 4⅝% Green Gilt 2037 is outside the window and should be excluded because the auction date was 2 June, even though settlement was 3 June. Inside the window, the currently scheduled conventional auctions are: 11 June 4% Treasury Gilt 2029 (maturity 22 May 2029, so short), 16 June 4⅞% Treasury Gilt 2036 (31 July 2036, medium), 24 June 4⅛% Treasury Gilt 2031 (7 March 2031, short), 2 July 4⅝% Green Gilt 2037 (7 March 2037, medium), 7 July 4⅛% Treasury Gilt 2033 (7 March 2033, short), 16 July 4⅞% Treasury Gilt 2036 (medium), 21 July 4% Treasury Gilt 2029 (short), and 4 August the new conventional gilt maturing 7 March 2032 (short). That is five short auctions and three medium auctions before any sizing assumptions are made. (dmo.gov.uk)
The broader issuance backdrop points to a relatively low medium share. In the revised 2026-27 remit, the DMO planned £95.0 billion of short conventional auction sales versus £56.0 billion of medium conventional auction sales, across 20 short and 15 medium auctions respectively. That implies average planned auction cash of about £4.75 billion for short versus about £3.73 billion for medium. The remit also says green gilt issuance of £12.0 billion cash is planned in 2026-27, subject to demand and market conditions, and that auction sizes are set case-by-case. Then, in the final July-September calendar, the DMO ended up with six short and only four medium conventional auctions, while a previously proposed long auction and a previously proposed medium auction were shifted into tenders rather than auctions. Because the resolution counts auctions only, that final calendar composition mechanically pulls the expected medium share down. (dmo.gov.uk)
For sizing, I anchor on recent DMO results for the same lines. Recent comparable cash raised were £4.966 billion for the 21 April 2029 auction, £4.719 billion for the 12 May 2031 auction, £3.952 billion for the 21 May 2036 auction, and £3.901 billion for the 27 May 2033 auction. The 2 June 2037 green auction raised about £3.156 billion cash on £3.25 billion nominal and had no PAOF. For the two nearest in-scope auctions, the DMO has already announced £5.0 billion nominal for the 11 June 2029 auction and £4.25 billion nominal for the 16 June 2036 auction. (dmo.gov.uk)
The green line is the biggest reason I do not push the medium share higher. The March 2026 syndication of 4⅝% Green Gilt 2037 raised about £6.201 billion cash, and the 2 June auction added about £3.156 billion, so roughly £9.357 billion of the £12.0 billion planned green total had already been used before the 2 July auction. That leaves only about £2.643 billion of room against the stated green plan. This does not force the DMO to size the 2 July auction at that exact amount, because the remit explicitly leaves room to respond to demand and market conditions, but it does make a repeat of another large green auction less likely in my base case. This is an inference from the official plan plus the official results, not an announced DMO decision. (dmo.gov.uk)
My base case therefore assumes roughly the following cash raised, in £ billions: 11 Jun 2029 = 5.05, 16 Jun 2036 = 4.20, 24 Jun 2031 = 4.45, 2 Jul Green 2037 = 2.70, 7 Jul 2033 = 4.15, 16 Jul 2036 = 4.05, 21 Jul 2029 = 5.00, and 4 Aug new 2032 = 4.90. That gives a medium-auction share of 31.739%. I also checked lower-medium and higher-medium scenarios and averaged them; the weighted scenario mean is 31.911%. So I center the forecast at about 31.9%. Most of the uncertainty is from the eventual size of the 2 July green auction, PAOF take-up on short auctions, the launch size of the new 2032 gilt, and the small residual risk that the DMO alters the calendar or sizes before resolution. (dmo.gov.uk)
I used only official DMO materials. The April-to-June 2026 calendar fixes conventional auctions on 11 June (4% Treasury Gilt 2029), 16 June (4⅞% Treasury Gilt 2036) and 24 June (4⅛% Treasury Gilt 2031), while the July-to-September 2026 calendar fixes 2 July (4⅝% Green Gilt 2037), 7 July (4⅛% Treasury Gilt 2033), 16 July (4⅞% Treasury Gilt 2036), 21 July (4% Treasury Gilt 2029) and 4 August (a new conventional gilt maturing on 7 March 2032). The 3 June and 12 August auctions are index-linked and therefore excluded from this question. (dmo.gov.uk)
The key subtlety is that the resolution buckets are based on residual maturity at issuance, not on how market participants loosely talk about the bond. Using DMO maturity dates—22 May 2029 for 4% Treasury Gilt 2029, 31 July 2036 for 4⅞% Treasury Gilt 2036, 7 March 2031 for 4⅛% Treasury Gilt 2031, 7 March 2037 for 4⅝% Green Gilt 2037, 7 March 2033 for 4⅛% Treasury Gilt 2033, and 7 March 2032 for the new August gilt—the residual maturities for the eight included auctions are about 2.92, 10.08, 4.75, 10.67, 6.67, 10.00, 2.83 and 5.58 years respectively. So only three auctions fall in the Medium bucket: 16 June 2036, 2 July 2037, and 16 July 2036. The 7 July auction of 4⅛% Treasury Gilt 2033 is short, not medium, because it matures on 7 March 2033. That one classification choice is the most important driver of the forecast. (dmo.gov.uk)
The broader issuance backdrop also points to a relatively low medium share. In the revised 2026-27 remit, planned auction issuance is £95.0 billion for short conventional gilts versus £56.0 billion for medium conventional gilts. More usefully for this forecast, the DMO’s current-remit table as of 9 June 2026 shows auction sales still required of £76.988 billion in the short bucket and £48.892 billion in the medium bucket, with 16 short auctions and 12 medium auctions remaining. That implies roughly £4.81175 billion of cash per remaining short auction versus £4.07433 billion per remaining medium auction. Applying those remaining-per-auction averages to this window’s five short and three medium auctions gives an even-flow estimate of 100 × (3 × 4.07433) / (5 × 4.81175 + 3 × 4.07433) = 33.6888%. (dmo.gov.uk)
I cross-checked that top-down estimate against recent auction cash results for the same lines. Official DMO auction data through 9 June show total cash raised of £4.96577 billion for the 21 April auction of 4% Treasury Gilt 2029, £4.71919 billion for the 12 May auction of 4⅛% Treasury Gilt 2031, £3.90148 billion for the 27 May auction of 4⅛% Treasury Gilt 2033, £3.95249 billion for the 21 May auction of 4⅞% Treasury Gilt 2036, and £3.15591 billion for the 2 June auction of 4⅝% Green Gilt 2037. A bottom-up window estimate built from those anchors again lands in the low-to-mid 30s, roughly 33% to 35%, depending mainly on the still-unannounced sizes of the later July/August auctions and on PAOF usage. (dmo.gov.uk)
PAOF is the main source of skew because the DMO’s auction totals include it. The DMO’s 2024-25 Annual Review says PAOF was activated in whole or in part at 44 of 77 auctions, raising £19.4 billion. In 2026-27-to-date results, short auctions have already seen sizable PAOF take-up (for example on 9 April 2033 and 12 May 2031), while the only non-green medium auction so far (21 May 2036) did not. I therefore infer a mild downside tilt for the medium share relative to a simple no-PAOF model, though only a mild one because future PAOF take-up is uncertain and the green 2037 auction has no PAOF anyway. (dmo.gov.uk)
My final forecast is 33.8%. I think the most likely resolution range is fairly tight because the auction identities in the window are mostly fixed already and the biggest remaining uncertainties are sizes, prices/cash proceeds, and PAOF take-up rather than which bonds will be sold. My rough 80% range is about 31.0% to 36.8%. (dmo.gov.uk)
The DMO calendars now make the composition of the window mostly knowable. Between Tuesday 3 June 2026 and Tuesday 11 August 2026 inclusive, the outright auction dates in scope are: 11 June (4% Treasury Gilt 2029), 16 June (4⅞% Treasury Gilt 2036), 24 June (4⅛% Treasury Gilt 2031), 2 July (4⅝% Green Gilt 2037), 7 July (4⅛% Treasury Gilt 2033), 16 July (4⅞% Treasury Gilt 2036), 21 July (4% Treasury Gilt 2029), and 4 August (new conventional gilt maturing 7 March 2032). The 3 June and 1 July auctions are index-linked and excluded, while the next index-linked auction on 12 August is outside the window. So, absent a schedule change, the denominator is eight conventional auctions. (dmo.gov.uk)
Using the resolution formula with redemption months from the prospectuses and calendar, the medium bucket contains only the two 4⅞% Treasury Gilt 2036 auctions and the 4⅝% Green Gilt 2037 auction. 4⅞% Treasury Gilt 2036 redeems on 31 July 2036; 4⅝% Green Gilt 2037 redeems on 7 March 2037; 4⅛% Treasury Gilt 2031 redeems on 7 March 2031; 4⅛% Treasury Gilt 2033 redeems on 7 March 2033; and the new August gilt matures on 7 March 2032. That gives residual maturities of about 10.1 years on 16 June 2026, 10.7 years on 2 July 2026, and 10.0 years on 16 July 2026 for the medium auctions, versus about 2.9 years for 11 June 2029, 4.8 years for 24 June 2031, 6.7 years for 7 July 2033, 2.8 years for 21 July 2029, and 5.6 years for 4 August 2032, all short. There are no long conventional auctions in the window. (dmo.gov.uk)
My first quantitative anchor is the DMO’s current-remit arithmetic. As of 9 June 2026, the DMO shows £76.988bn of short auction sales and £48.892bn of medium auction sales still required to meet the auction plans, with 16 short and 12 medium auctions remaining. That implies average remaining auction cash of about £4.81bn for short versus £4.07bn for medium; applying those averages to the five short and three medium auctions inside this window gives a base-case medium share of 33.689%. (dmo.gov.uk)
My second anchor is year-to-date realized auction cash. The same current-remit update shows short auction proceeds to date of £16.687bn and medium auction proceeds to date of £7.108bn. Given the auction calendar up to 9 June 2026, that corresponds to four completed short auctions and two completed medium auctions in 2026-27, or average realized cash of about £4.17bn per short auction and £3.55bn per medium auction. Applying those realized averages to a five-short/three-medium window gives 33.825%. (dmo.gov.uk)
My third anchor is a bottom-up size view. One short auction is already fixed at £5.0bn nominal on 11 June 2026. Recent comparable auctions were £4.0bn nominal for 4⅞% Treasury Gilt 2036 on 21 May and £3.25bn nominal for 4⅝% Green Gilt 2037 on 2 June; recent short auctions around the current benchmarks were £5.0bn for 4% Treasury Gilt 2029 on 21 April, £4.0bn for 4⅛% Treasury Gilt 2031 on 12 May, and £4.0bn for 4⅛% Treasury Gilt 2033 on 27 May. Auction prices across these short and medium issues have all been clustered around the high-97 to high-99 area, so cash-share differences versus nominal-share differences should be small. A representative granular scenario therefore lands near 33.066% medium cash share. (dmo.gov.uk)
Blending those three anchors with extra weight on the remit-based arithmetic produces 33.543% before judgmental adjustment. I then shade very slightly lower rather than higher because medium issuance has already been front-loaded via syndication: the current-remit table shows medium syndication sales to date of £23.829bn against a planned £20.0bn, which reduces the pressure to run medium auctions materially above trend later in the summer, even though the auction calendar itself is mostly fixed. My point forecast is therefore 33.5%. I put most of the probability mass roughly in the 30% to 37% area; the main uncertainty is the exact auction sizes still to be announced one week before the remaining operations, not which bucket the scheduled auctions belong to. (dmo.gov.uk)
My forecast starts from the official DMO auction calendars that are already published as of Tuesday, June 9, 2026. In the resolution window from June 3, 2026 through August 11, 2026 inclusive, the scheduled conventional auctions are: June 11 (4% Treasury Gilt 2029), June 16 (4⅞% Treasury Gilt 2036), June 24 (4⅛% Treasury Gilt 2031), July 2 (4⅝% Green Gilt 2037), July 7 (4⅛% Treasury Gilt 2033), July 16 (4⅞% Treasury Gilt 2036), July 21 (4% Treasury Gilt 2029), and August 4 (new conventional gilt maturing on 7 March 2032). The June 3 and July 1 auctions are index-linked and therefore excluded under the question rules; the August 12 auction is also index-linked and in any case falls after the window ends. (dmo.gov.uk)
Using the DMO maturity dates, those eight qualifying auctions split cleanly into five Short and three Medium auctions. Short: June 11 2029 gilt (matures 22 May 2029), June 24 2031 gilt (7 March 2031), July 7 2033 gilt (7 March 2033), July 21 2029 gilt, and August 4 new gilt maturing 7 March 2032. Medium: June 16 2036 gilt (31 July 2036), July 2 green 2037 gilt (7 March 2037), and July 16 2036 gilt. No scheduled long conventional auction falls inside the window. (dmo.gov.uk)
A good baseline is the DMO’s current remit progress sheet published for June 9, 2026. It shows auction sales still required to meet plan of £76,988 million in Short and £48,892 million in Medium, with 16 Short and 12 Medium auctions remaining. That implies average remaining auction cash of about £4.812 billion per Short auction and £4.074 billion per Medium auction. If I simply apply those category averages to this window’s five Short and three Medium slots, the implied Medium share is 33.689%. (dmo.gov.uk)
I then refine that baseline with stock-specific information. Two relevant operations are already announced in nominal size: the June 11, 2026 4% Treasury Gilt 2029 auction is £5.0 billion nominal, and the June 16, 2026 4⅞% Treasury Gilt 2036 auction is £4.25 billion nominal. Recent accepted average clean prices for the same or directly relevant gilts were 99.315 for 4% Treasury Gilt 2029 (April 21, 2026), 98.812 for 4⅞% Treasury Gilt 2036 (May 21, 2026), 97.747 for 4⅛% Treasury Gilt 2031 (May 12, 2026), 97.537 for 4⅛% Treasury Gilt 2033 (May 27, 2026), and 97.105 for 4⅝% Green Gilt 2037 (June 2, 2026). Those price anchors matter because the question resolves on cash raised, not nominal amounts. (dmo.gov.uk)
For the six qualifying auctions whose nominal sizes are not yet announced, I use conservative central assumptions close to recent DMO practice: roughly £4.0 billion nominal for the June 24 2031 re-opening, £3.25 billion for the July 2 green 2037 re-opening, about £4.0 billion for the July 7 2033 re-opening, about £4.25 billion for the July 16 2036 re-opening, about £5.0 billion for the July 21 2029 re-opening, and about £4.5 billion for the August 4 launch of the new March 2032 gilt (with that new launch modeled near par). Plugging those assumptions into recent clean-price anchors gives an instrument-specific estimate of 34.188%. (dmo.gov.uk)
I do not want to overweight that single sizing path, so I blend it with the more mechanical remit-average baseline. The simple blend of 33.689% and 34.188% is 33.939%, which I use as my central forecast.
For uncertainty, the key point is that the bucket classification itself is now mostly locked in by the published calendar and maturity dates; almost all residual uncertainty comes from how much cash each qualifying auction raises. My downside scenario is a relatively small July 2 green auction, a slightly smaller July 16 medium auction, and somewhat larger short issuance on June 24 and August 4; that gives about 31.75%. My upside scenario is the opposite mix, which gives about 36.17%. I then allow a small extra tail for the low-probability chance of schedule changes or unusually atypical sizing. (dmo.gov.uk)
Bottom line: I expect the answer to resolve in the mid-30s, with the most likely outcome very close to 34%. My point forecast is 34.0%.