The recent pullback in gold cannot be explained by the level of real interest rates alone. Two more critical lines of analysis are at play: one is dollar credibility, tied to US fiscal pressures and central bank allocation demand; the other is its financial nature, tied to global liquidity, dollar exchange rates, and safe-haven trading. The current issue is that neither of these factors is supportive of gold.
A core conclusion from a research report dated July 1st is that the monetary and financial attributes are the two most significant drivers of the gold price, and both are currently facing headwinds, suggesting the correction trend may persist in the near term.
This analytical framework incorporates AI capital expenditure into gold pricing. The logic is straightforward: if capital spending by tech giants supports the US economy, the US government may not need to rely on larger fiscal deficits to underpin growth. This would ease market concerns over US fiscal sustainability, weakening the medium-term allocation case for gold. Simultaneously, a strong economy with sticky inflation would limit the Federal Reserve's room for interest rate cuts, potentially even strengthening expectations for hikes, thereby pressuring gold's financial appeal.
Therefore, for gold to regain a period of opportunity, one need not wait solely for geopolitical risks or a decline in US real yields. A closer observation window might be the July-August US earnings season. With prices for components like memory having risen sharply, if tech giants do not correspondingly and significantly revise their AI capital expenditure plans upward, it implies a reduction in actual construction scale. This could serve as a signal for gold to stabilize.
Gold Price is Not Solely Priced by Real Rates, a Lesson from 2023
In recent years, a common misjudgment regarding gold has been the tendency to simplify it as merely the inverse of real interest rates. This explanation, while effective in many periods, is incomplete.
2023 served as a counterexample. The Federal Reserve was still in a tightening cycle, and US real rates were elevated. According to the traditional framework, gold should not have been strong. Yet the gold price clearly diverged from real rates, driven by US fiscal policy: despite the rate-hiking cycle, the US government pursued significant fiscal expansion. Rising debt servicing pressures damaged dollar credibility, providing support for gold from allocation flows.
The fourth quarter of 2025 presented another type of dislocation. The US fiscal deficit contracted, central bank gold purchases declined, weakening the marginal allocation case. However, intensifying expectations for Fed rate cuts and substantial inflows into gold ETFs saw trading capital take the lead, driving the gold price higher. This illustrates that gold can be propelled by both "dollar credibility" concerns and "liquidity trading."
What Long-Term Allocation Funds Watch: Fiscal Sustainability and Central Bank Faith in the Dollar
The research framework clearly divides factors influencing the gold price into two categories: one concerning long-term allocation logic, and the other concerning short-term trading sentiment.
The core of allocation logic lies in the US government's debt-servicing capacity and the choices of global central banks, specifically two indicators:
US Fiscal Pressure: Larger fiscal deficits and a higher proportion of government interest payments to revenue increase market worries about US debt sustainability, damaging dollar credibility and highlighting gold's allocation value.
Global Central Banks' Stance on the Dollar: If central banks worldwide are reducing dollar-denominated assets and increasing gold holdings, they are voting with real money, expressing distrust in the dollar system. Once this force becomes a trend, it provides long-term support for gold prices.
This explains the behavior of long-term allocation capital: it may not chase short-term trends, but once a directional view forms, it can often support gold prices through a sustained trend.
What Short-Term Trading Funds Watch: Dollar Direction, Safe-Haven Currencies, and Global Liquidity Conditions
Gold's financial nature speaks more to the language of short-term trading funds. The core of trading logic revolves around market liquidity and risk aversion, monitored through three signals:
First, safe-haven currencies like the Japanese Yen and Swiss Franc. Sharing gold's safe-haven attributes, their movements can reflect market risk sentiment more accurately than the US Dollar Index at times.
Second, global liquidity. As a non-yielding asset, gold sees increased buying interest when liquidity is ample. This liquidity factor is synthesized from indicators like global central bank policy expectations and global leverage levels.
Finally, financial stress. The OFR Financial Stress Index is used as a proxy for global financial system pressure. When this index is above zero, gold prices are often more prone to rise. This isn't the entirety of gold's long-term pricing, but during risk event shocks, it can create a pulse-like impact.
Trouble Arises Only When Both Lines Face Headwinds
Backtesting results deliver a clear conclusion: gold does not require both the allocation and trading factors to turn positive simultaneously. As long as one is activated, its performance is typically not poor.
In the sample from January 2007 to June 2026, gold's cumulative return was highest when both factors were positive. However, when the allocation factor was positive and the trading factor negative, the monthly win rate was still 60.3%. When the allocation factor was negative and the trading factor positive, the monthly win rate was even higher at 76.2%.
The truly unfavorable state is when both factors are negative. This occurred in 62 months within the sample, with an average monthly annualized return of -8.6% and a monthly win rate of only 41.9%, below the full-sample win rate of 57.1%.
Based on this, a simple integrated timing strategy emerges: hold gold when either the allocation or trading factor is greater than zero; hold no position when both are below zero. From January 2007 to June 2026, this strategy yielded an annualized return of 13.1%, compared to 9.9% for London gold. It effectively avoided sustained drawdowns in periods like 2013-2015 and 2022 while participating in several gold bull markets.
However, a boundary must be clarified: looking at the allocation or trading factor alone can capture parts of the upside but cannot outperform the benchmark. The difficulty with gold lies precisely in its frequent pricing by different types of capital in different phases, taking turns to drive the market.
Gold's Current Problem: AI Enables a US Economy That Doesn't Rely on Fiscal Support
Since March 2026, the model indicates that both the gold allocation and trading factors have turned negative. In other words, both monetary and financial attributes are facing headwinds.
The key variable here is AI capital expenditure.
From 2020 to 2025, US economic performance showed a clear correlation with fiscal deficit trends, with fiscal expansion being a key driver supporting economic strength. However, since 2026, a divergence has appeared: US economic activity has improved while the fiscal deficit has been contracting. This suggests US growth is increasingly driven by the private sector rather than the government.
Breaking it down, domestic private investment has become the most crucial factor pulling US GDP growth. With tech giant capital expenditure supporting economic growth, the US government does not need to continue relying on strong fiscal stimulus to maintain growth resilience. For gold, this weakens the logic chain of "unsustainable US fiscal policy -> damaged dollar credibility -> rising gold allocation value."
On the other side, AI capital expenditure supporting the economy also narrows the Federal Reserve's room for rate cuts. A strong economy coupled with stubborn inflation raises expectations for rate hikes or at least suppresses expectations for cuts, naturally putting pressure on gold as a non-yielding asset.
Gold's Potential Phased Opportunity May Await a Peak in AI Capital Expenditure Expectations
If AI capital expenditure is what's pressuring gold, then a signal for gold's stabilization might also come from a shift in AI capital expenditure expectations.
The nearest observation window is the July-August US earnings season. With significant price increases recently in segments like memory, if tech giants do not correspondingly and substantially revise their AI capital expenditure plans upward, it implies that actual construction scale may decline in the face of rising costs.
Once the market begins to trade on the theme of "AI capital expenditure peaking for a phase," the narrative of the US economy being supported by private investment would be re-examined, and the Federal Reserve's room for rate cuts might reopen. For gold, this could potentially bring about a phased allocation opportunity.
The risk lies in the fact that these conclusions are derived from historical data and model processing. Historical backtesting does not guarantee future results, and parameter settings can influence outcomes. Gold prices themselves exhibit significant volatility, and both factors turning negative does not equate to a one-way decline for gold. However, it at least indicates that a short-term reversal would require a stronger catalyst.
Comments