Gold structural debasement bid
What changed
The headline signal — how confidence moved from the previous snapshot, and why.
Stepped 0.85 to 0.84 and widened 0.05 to 0.06 (from 2.2.3). A single yield-driven down session is named noise; what moved the estimate is the PATTERN - two consecutive windows of hard drawdowns (spot ~$4,090, cumulatively ~10% off the late-May band,
GDX -6.4% to 73.81) in which gold repeatedly failed to catch a safe-haven bid through major escalations, because
the 4.2% CPI and
the ~4.54% 10Y keep the yield channel dominant. The structural debasement case is untouched - no leg of the AndCondition triplet is in motion (durable peace moved away this window) and the central-bank-buying supports stand - so the mean steps only -0.01; the band widens because the near-term expression of the thesis keeps losing to rates, and the effective evidence weight for the CURRENT price path is genuinely lower. Beta(30.2, 5.8) ~36 effective observations. The horizon reflects the structural character: the AndCondition triplet realistically takes a year-plus to assemble.
The thesis
The claim and where confidence stands now.
Gold sits in mid-cycle of a multi-year structural bull market driven by central-bank buying, sovereign de-dollarization, $39T US debt, stock-bond correlation breakdown, and Fed independence concerns. Q1 2026 set records on multiple dimensions - LBMA quarterly average $4,873/oz, central-bank net purchases 244 tonnes (highest Q1 ever, +17% QoQ), aggregate Q1 demand value $193B, bar-and-coin demand 474 tonnes (+42% YoY, second-highest quarterly figure on record). After the Monday US cash recovery (GLD 418.43 +0.27%, spot back into the $4,565-$4,570 band), Tuesday morning spot is range-bound $4,531-$4,570 - Asian-session $4,531 low recovering through the European session on dollar weakness driven by the Iran-rejection-but-no-strike framing. Physical demand strong - Shanghai-London spot differentials remain positive. But the 10Y rose to a 4.62% Tuesday close (intraday high 4.67%) - the operative tactical headwind reasserting - and gold pulled back below the band: spot ~$4,486.86,
GLD 411.50 (-1.66%),
GDX 83.78 (-3.86%); Wednesday pre-market spot steadied near $4,498.90 (-0.27%). The structural-bull supports are intact - JPM $5,055-$6,300, UBS $6,200, Citi $5,000-$7,000 targets; 2026 CB projection at 755 tonnes; 95% CB survey response confirming gold-reserve intentions. AndCondition invalidation still requires three hard things simultaneously - durable US-Iran peace AND Fed credibility restored AND sustained 6-month deficit decline - the
Senate's war-powers advance moves further away from the first leg, but the day's action is a tactical price pullback, not a structural break. Update May 20: the pullback reversed -
GLD 417.40 (+1.43%),
GDX 86.36 (+3.08%) - as
yields eased, the Tuesday tactical headwind flipping to a tailwind; the
Iran talks entering final stages are a small structural offset toward the durable-peace invalidation leg. Update May 21: two tactical forces offset -
Treasury yields rebounded (headwind) while
US-Iran deal optimism faded (safe-haven tailwind) - leaving the structural-bull read intact. Update May 22: the two tactical forces offset again with signs flipped -
yields eased Friday (tailwind) while the
leaked draft ceasefire revived peace optimism (safe-haven headwind) - and Thursday closed flat (
GLD 416.99,
GDX 85.99). Update May 25: gold gave back a little into the Friday close -
GLD 413.82 (-0.76%),
GDX 85.02 (-1.13%) - as the revived peace optimism outweighed the easing-yield tailwind on the day; a tactical pullback, with the structural-bull supports unchanged over a quiet holiday weekend. Update May 26: gold firmed over the holiday -
spot advanced 1.18% to $4,562.69 Monday on dollar softness and a safe-haven bid amid the
fresh US strikes on Iran, with
the 10Y easing to ~4.51% a tailwind - recovering the Friday give-back on unchanged structural supports. Update May 27: gold gave Monday's pop back on the first cash session -
spot fell 1.74% to $4,489.65 Tuesday on risk-on / deal optimism - though
GLD closed net flat at 414.00 and
GDX ran +4.09% to 88.50 with
the 10Y easing to ~4.48% a tailwind; a tactical pullback on unchanged structural supports. Update May 28: the Wednesday cash close brought a sharper selloff -
spot gold fell ~1.67% to $4,438.98 on risk-on flows and dollar firmness as Iran-deal optimism overweighed the easing 10Y - with
GLD 408.49 (-1.33%) and
GDX 85.44 (-3.46%) confirming the tactical pullback. Structural supports (244t Q1 central-bank buying, JPM/UBS/Citi targets, 755t 2026 projection) are unchanged. Update May 29: bullion rebounded on the otherwise risk-on day -
spot recovered to ~$4,470,
GLD 412.77 (+1.05%),
GDX 87.18 (+2.04%) - with
the 10Y easing to 4.46% a tailwind; the in-line April PCE and the ~20% monthly oil collapse point to easier forward inflation while the structural debasement supports are unchanged, and the AndCondition invalidation triplet is no closer. Update June 1: the rebound extended into the Friday close -
GLD 417.12 (+1.05%) and
GDX 89.49 (+2.65%), the miners leading for a second consecutive day - and the weekend
stall of the US-Iran talks plus continuing blockade enforcement restores the safe-haven tailwind that the peace-MoU optimism had been eroding; the structural supports and the AndCondition invalidation triplet (which requires durable peace as its FIRST leg, now further away) are unchanged. Update June 2: the two-day rebound broke on the risk-on record session -
spot gold fell ~1.9% Monday to ~$4,455,
GLD 411.26 (-1.40%),
GDX 86.68 (-3.14%) (the miners giving back their Friday leadership) - as the equity records and
Trump's peace framing outweighed
the Iran-talks collapse on the day, with
the 10Y rising to ~4.47% an added headwind. Tuesday spot is
recovering to ~$4,514 (+$54). The structural supports (244t Q1 central-bank buying, JPM/UBS/Citi targets, 755t 2026 projection) are unchanged, and the talks collapse moves the durable-peace leg of the AndCondition further away. Update June 3: the recovery completed into the Tuesday close -
spot gold closed ~$4,530 (+1.7%),
GDX +1.58% to 88.05 (the miners reclaiming leadership),
GLD +0.17% to 411.95 - recovering the full Monday risk-on give-back, with
the 10Y easing to ~4.43% a tailwind. The diplomatic reversal (
the memorandum pending Trump's signature) is a small move TOWARD the durable-peace leg of the AndCondition triplet, but the other two legs (Fed credibility restored; six months of deficit decline) remain untouched, and the structural supports are unchanged. Update June 4: gold took a tactical hit on the broad risk-off day -
spot fell ~1.6% to ~$4,457,
GDX -3.5%,
GLD -1.0% - failing to catch a safe-haven bid despite
the Iranian strikes on Kuwait and Bahrain, a single down session that trips none of the AndCondition triplet; the structural debasement supports are unchanged. Update June 5: gold fell again with the yield spike -
spot ~$4,327 (-2.9%),
GLD to 396.24,
GDX -5.9% to 78.84 - a yield-driven tactical drawdown that still trips none of the AndCondition triplet; the central-bank-buying support stands. Update June 11: the drawdown deepened materially and the haven test failed again - through the sharpest war escalation since February (
direct Iran-Israel ballistic exchanges, US strike waves), gold FELL:
spot ~$4,090 late Wednesday (readings $4,079-$4,146 through the day), down ~5.5% from June 5,
GLD 374.58,
GDX -6.4% to 73.81 - as
the 4.2% CPI and
the ~4.54% 10Y kept the yield channel dominant and risk-off margin flows sold the metal with everything else. The cumulative drawdown from the late-May ~$4,530 band is ~10%. The AndCondition triplet is untouched (durable peace moved AWAY; no Fed-credibility restoration; no deficit decline) and the structural supports (244t Q1 central-bank buying, 755t 2026 projection, JPM/UBS/Citi targets) stand.
Drivers
The underlying macro forces this thesis expresses - the loading mean is how much each force drives the thesis, the stddev our confidence in the mapping.
Supporting evidence
Typed, citation-backed observations across time, grouped by strength. Hover a point for the claim.
What would invalidate this
The machine-evaluable conditions that would falsify the thesis.