Central bank gold purchasing and the gold price outlook Report Interpretation

Overview

This commodities comment examines whether central-bank gold demand remains strong and what it means for Goldman Sachs' end-2026 gold outlook. The institution finds that official purchases remain elevated, providing a foundation for its $4,900/toz forecast, but highlights materially wider upside and downside paths driven by ETF demand, Fed expectations and options-market hedging.

Core views

Goldman Sachs estimates that central banks purchased 44 tonnes of gold in July, compared with a pre-2022 average of 17 tonnes. China accounted for 35 tonnes and was the largest identifiable buyer. On a three-month seasonally adjusted basis, the institution estimates the purchasing trend at 91 tonnes per month, supporting its view that official-sector demand remains structurally strong. The reported 44-tonne July nowcast may understate recent buying. Goldman Sachs' London OTC measure tracks gold moving into domestic storage or third-party custodians such as the BIS in Switzerland, but excludes changes in foreign central-bank gold held at the Bank of England. Bank of England custody holdings rose by 63 tonnes in July. Although some of that increase likely reflected transfers from the New York Fed, the Bank of England increase more than offset the New York decline, which Goldman Sachs says indicates additional central-bank purchases may not be captured in its July estimate. Goldman Sachs maintains a fair-value forecast of $4,900/toz by end-2026 and sees net upside risk to that level. Its baseline assumes continued strong official demand, averaging 50 tonnes per month in 2026 and 40 tonnes per month in 2027, together with a recovery in private-investor ETF demand while the Fed remains on hold in 2026. The base case does not include elevated demand for macro-policy hedges through gold call options. If ETF inflows recover while the current elevated call-option positioning persists, dealer hedging could mechanically amplify a rally and drive gold materially above the forecast. The institution also outlines a downside policy scenario. A renewed rise in expectations for Fed hikes could prompt dealer hedge unwinds and an unusually sharp correction. If the Fed actually hikes, demand for gold as a macro-policy hedge could partially unwind as concerns about developed-market central-bank independence ease; rate-sensitive ETF investors could also sell into higher rates. In that scenario, Goldman Sachs estimates gold could reach $4,440/toz by end-2026—well below its $4,900/toz baseline, though still slightly above current levels because continued central-bank buying would eventually more than offset part of the pressure. Outside the baseline, speculative positioning could build ahead of the US midterm elections and then reverse sharply once election-result clarity emerges. Goldman Sachs characterizes gold as a possible temporary “waiting room” allocation during high uncertainty, citing prior behavior before the 2016 Brexit referendum and the 2024 US presidential election. The latest available positioning data, as of September 1, 2026, were 438 tonnes, at the 67th percentile of the sample since 2014. Over the medium term, however, the institution believes forecast risks remain tilted upward because gold's share of private portfolios remains low and geopolitical developments, including Iran and broader tensions, could broaden diversification demand from central banks to private investors, including through concerns about G10 fiscal sustainability.

Analysis framework

Goldman Sachs combines a nowcast of central-bank purchases with custody-flow evidence to assess official demand. It then builds an end-2026 fair-value path from assumed central-bank buying, ETF demand and Fed policy, and contrasts it with upside and downside scenarios involving gold call-option positioning, dealer hedging, rate-sensitive ETF flows and speculative positioning around major political events.

Methodology notes

• Industry AnalysisSupply-demand framework

  Gold demand assessment using central-bank purchases, private-investor ETF flows and macro-policy hedge demand.

  The report treats sustained official buying and a potential recovery in ETF demand as demand supports for gold, while weaker hedge demand and ETF selling under higher rates create a downside path.

• Event-Driven and Behavioral FinanceEvent-driven analysis

  Scenario analysis of Fed decisions and high-uncertainty political events.

  Goldman Sachs examines how a Fed hike or the resolution of election uncertainty could change gold demand, positioning and dealer hedging, producing sharp price moves.

• Other

  Goldman Sachs central-bank gold-purchase nowcast based on London OTC and custody flows.

  The nowcast estimates gold moving into domestic storage or third-party custodians, but excludes foreign central-bank holdings at the Bank of England, so the reported July estimate may not capture all buying.

Asset mapping & comparison

Structured mapping from thesis to named assets (strengths, weaknesses, peers, risks).

• Gold

  The report's primary commodity subject; sustained central-bank purchases support the base-case price path.

  Strengths

  Official demand remains far above the pre-2022 average, and potential private-investor diversification could add demand.

  Weaknesses

  Gold is exposed to changes in Fed-hike expectations, rate-sensitive ETF selling and post-event speculative-position unwinds.

  Comparison

  The July central-bank nowcast of 44 tonnes compares with a pre-2022 average of 17 tonnes; the three-month seasonally adjusted trend is 91 tonnes per month.

  Risks

  A Fed-hike scenario could take gold to $4,440/toz by end-2026; elevated call-option positioning can also increase two-sided volatility.

Key data

• Estimated central-bank gold purchases44 tonnes in JulyGoldman Sachs nowcast; versus a pre-2022 average of 17 tonnes.
• Three-month seasonally adjusted central-bank purchase trend91 tonnes per monthGoldman Sachs estimate through July.
• China's identifiable July purchases35 tonnesChina was the largest identifiable buyer.
• Bank of England custody holdings change+63 tonnes in JulyThe increase more than offset the decline at the New York Fed and may signal purchases omitted from the July nowcast.
• Base-case gold forecast$4,900/toz by end-2026Assumes average central-bank purchases of 50 tonnes per month in 2026 and 40 tonnes per month in 2027, plus a recovery in ETF demand.
• Fed-hike scenario$4,440/toz by end-2026Would reflect hedge-demand unwinding, dealer hedge unwinds and rate-sensitive ETF selling.
• Latest speculative positioning438 tonnes; 67th percentileLast available data as of 2026-09-01, in a sample since 2014.

Impact & implications

The report argues that persistent central-bank buying provides an important demand floor for gold and supports its end-2026 base case. However, price outcomes may be amplified in either direction by ETF flows and dealer hedging linked to call-option demand, while Fed policy expectations and post-election position unwinds could cause sharp corrections.

Risks

• A renewed rise in Fed-hike expectations could trigger dealer hedge unwinds, reduce macro-policy hedge demand and prompt ETF selling.
• Speculative positioning may rise ahead of the US midterm elections and reverse sharply after election-result clarity.
• The July central-bank-purchase nowcast may omit purchases reflected in foreign central-bank custody holdings at the Bank of England.

What to watch

• Central-bank purchase volumes, particularly China's buying and custody-flow developments at the Bank of England and New York Fed.
• Whether private-investor gold ETF inflows recover while the Fed remains on hold.
• Gold call-option positioning and the potential for dealer hedging to amplify price moves.
• Fed-hike expectations and their effect on hedge demand and rate-sensitive ETF holders.
• Speculative gold positioning ahead of and after the US midterm elections.