The PSEi-ICTSI Show, Part III: The Fulcrum Cracks—Benchmark-ism Unravels

 

You can't do well in investments unless you think independently. And the truth is, you're neither right nor wrong because people agree with you. You're right because your facts and your reasoning are right. In the end, that's all that counts. And there wasn't any question about the facts or reasoning being correct—Warren Buffett 

In this issue: 

The PSEi-ICTSI Show, Part III: The Fulcrum Cracks—Benchmark-ism Unravels

I. Ghost Month, Real Losses

II. The ICTSI Fulcrum Cracks

III. Breadth Confirms the Divergence

IV. From Prop to Drag: The Mechanics of Concentration in Reverse

V. Rotation Without Broadening: Speculative Capital in Search of the Next Punt

VI. The Arrière-Pensée: Confidence Is the Cover Story, Collateral Is the Point

VII. Conclusion: Benchmark-ism Meets Its Own Arithmetic 

The PSEi-ICTSI Show, Part III: The Fulcrum Cracks—Benchmark-ism Unravels 

When one stock becomes the market, its reversal becomes the market's problem: August exposes the concentration, liquidity, and benchmark-ism beneath the PSEi's manufactured confidence 

In Part I we flagged ICTSI as the PSEi 30's single point of vulnerability. In Part II we showed how that vulnerability had metastasized into concentrated liquidity, shrinking participation, and outright benchmark-ism. August supplied the arithmetic Part II warned was coming. 

I. Ghost Month, Real Losses

Figure 1

August is conventionally excused as the market's "ghost month"—a superstition with Chinese roots, ironically applied by the establishment commentaries to a market that isn't Chinese. The historical record hardly justifies the fear: from 2010 through 2025, August produced gains in only six of sixteen years, with an average loss of a marginal -0.33%. Mostly noise. (Figure 1, upper window)  

This August wasn't noise. The PSEi 30 fell 4.49% month-on-month—the fourth-largest monthly loss since 2010—dragging the year-to-date return to roughly -1.60% and the year-on-year figure to -3.24%. (Figure 1, lower table)

Figure 2 

The damage was compressed: a single closing week's 4.52% plunge erased three months of the index's ICTSI-driven advance. (Figure 2, upper image) 

Ghost month or not, the mechanism behind the reversal is the one this series has tracked for some time, but which intensified last June: ICTSI had become the fulcrum propping up the PSEi 30, and a fulcrum that lifts a benchmark on the way up can just as easily accelerate its fall. 

The timing again matters. 

The final week's plunge did not arrive in isolation. 

On Friday, the USDPHP closed at Php 62.265, a record, extending Thursday's prior record-low close of Php 61.888. Markets were also absorbing the BSP's latest policy signals, and the 7-month record fiscal and trade deficits. (Figure 2, lower chart) 

The PSEi 30 fell in lockstep, down for a fourth consecutive session and closing below 6,000, erasing gains since June. 

A rate hike aimed at defending the currency and a benchmark sliding on the same day are not coincidences to be filed separately from the ICTSI story. They are another expression of the same "resistance to adjustment" pattern this (stagflation) series has tracked across fiscal, monetary, and FX-regime policy: administrative efforts to keep the peso, the yield curve, and the index from clearing at levels the underlying imbalances would otherwise dictate. 

When that resistance gives way on the currency side, it rarely leaves the equity side untouched—not when equity "confidence" was doing double duty as collateral support in the first place. 

II. The ICTSI Fulcrum Cracks 

The same stock that manufactured the June-July rally supplied most of August's decline. ICTSI fell 7.86% week-on-week and 8.1% month-on-month, accounting for roughly 46% and 48%, respectively, of the PSEi's weekly and monthly losses. 

This is simply capitalization-weighting running in reverse. The identical arithmetic that let one company's 57% first-half return mask a negative-6.8% average constituent return in Part II is now transmitting one company's reversal directly into the headline index. 

A benchmark built to amplify a single stock's ascent is, by the same construction, built to amplify its descent.

Figure 3

ICTSI's share of the PSEi's free-float market cap slipped to 25.58% by month's end, down from a peak of 27.37% in the first week of August. (Figure 3, upper pane)         

The combined weight of the top five constituents eased from July's peak of 56.12% to about 54.7%. 

Small retreats in percentage terms—large in what they signal: the "national team" support that Part II documented in the daily trading choreography appears to be losing its grip on the tape. 

III. Breadth Confirms the Divergence 

Beneath the headline, the market's breadth told a different story from the index. 

Only 10 of the PSEi's 30 constituents rose in August; the average constituent decline was roughly -2.92%—materially less than the benchmark's -4.49%, as ICTSI's outsized free-float market-cap weight amplified its decline in the headline index. (Figure 3, lower visual)

Figure 4

The divergence was even more pronounced during the final week's meltdown: only 7 of the 23 constituents that traded that week rose, while the average change was -2.78%—another example of how free-float market-cap weighting magnified the impact of ICTSI's decline on the benchmark. (Figure 4, upper diagram) 

Exchange-wide breadth stayed only marginally negative—1,783 advances against 1,818 declines—a spread nowhere near as lopsided as the ICTSI-driven index move would suggest. (Figure 4, lower image) 

That gap between a nearly even advance-decline line and a sharply negative benchmark is itself the tell: the index and the market it purports to summarize are increasingly telling two different stories, exactly as Part II's mid-year data showed in reverse. 

IV. From Prop to Drag: The Mechanics of Concentration in Reverse

Figure 5 

The unraveling wasn't confined to price. It showed up in the plumbing of trading activity too. ICTSI's % share of Main Board volume fell sharply from 28.23% in July to 21.91% in August as foreign flows shifted from net buying (Php 5.265B in July) to net selling (Php 1.59B in August). (Figure 5, upper graph) 

The same expansion in ICTSI's trading dominance that Part II attributed to both rising ICTSI activity and shrinking activity elsewhere is now reversing on both counts at once: as ICTSI volume recedes, there is no offsetting pickup in the rest of the market to cushion the loss of liquidity. 

The pattern increasingly resembles South Korea's KOSPI, where Samsung and SK Hynix now function as the index in practice. The difference is that Korea's concentration rides a global AI-hardware cycle with real earnings behind it; ICTSI's had no comparable confirmation from global port-operator peers, a point already established in Part II. (Figure 5, lower chart) 

Main Board volume fell 17.24% year-on-year to about Php 113.108 billion; aggregate turnover fell 4.5% to Php 142.976 billion. Cross trades made up 18.74% of Main Board volume, a four-month high—more of the exchange's already-shrinking activity conducted off the central order book, away from open price discovery. 

V. Rotation Without Broadening: Speculative Capital in Search of the Next Punt 

As the dominant ICTSI trade weakened, the other big caps failed to fill the gap—or the expected ‘rotation’ didn’t happen. Instead, trading activity shifted toward smaller, more speculative names—restructuring rumors, backdoor-listing chatter, M&A gossip—even as overall volume contracted. 

Near-even breadth on falling volume is not a healthy broadening of participation. It looks more like speculative capital searching for the next short-term story once the previous market leader loses momentum, not new capital entering the market on improved fundamentals. 

The extraordinary ICTSI ascent that carried the PSEi through June and July appears to have reached its inflection point. What follows this kind of top is rarely a broad-based handoff to the rest of the index; it's usually a scramble among smaller, thinner names while the index itself searches for a new prop. 

The progression is telling: from speculation to outright gambling, as capital searches less for value than for the next trade capable of replacing the momentum it has lost—more recipe for capital consumption. 

VI. The Arrière-Pensée: Confidence Is the Cover Story, Collateral Is the Point 

In Part II, we named this dynamic “benchmark-ism”: political and institutional narrative management that uses market and economic statistics to manufacture the appearance of stability—or to keep the Overton window of acceptable economic outcomes from shifting toward adjustment. 

August is the moment that narrative met its own arithmetic. 

Gains have always been projected as "confidence"—a rising PSEi as evidence of "resilience," useful for a leadership with sagging approval ratings and an economy absorbing the mounting tensions from the aftershocks of the Iran-war oil shock, EO 110's price-suppression architecture, and a cascade of BSP relief measures already tracked in this series. But confidence was never the only objective, and arguably not the primary one. 

Elevated equity prices are collateral. They support asset values on bank and corporate balance sheets, cushioning a financial system already leaning on the BSP's regulatory relief and foreign-exchange interventions to keep the credit cycle turning. 

Put bluntly: a rising PSEi, and specifically a rising ICTSI, was never just about optics. It was about propping up the collateral base that a decaying credit cycle depends on to keep rolling over. That is the arrière-pensée behind the applause Part II described—the second, unstated motive sitting behind the first. 

The trouble with using one stock as collateral infrastructure for an entire financial system is that the mechanism is symmetric. 

What inflates the collateral on the way up can deplete it on the way down, potentially at a leveraged pace 

August didn't just cost the PSEi 4.49%. It cost the system a slice of the very collateral cushion the whole exercise was designed to build. 

VII. Conclusion: Benchmark-ism Meets Its Own Arithmetic 

This isn't the first time a single Philippine equity story has been mistaken for—or dressed up as—a market. The DigiPlus [PSE: PLUS] gaming-stock bubble of 2025 ran on the same logic: a policy-fueled speculative vehicle inflated past any relationship to its fundamentals, celebrated rather than examined, until the arithmetic of momentum reversed on its own. 

We called that top in real time and traced its roots back to the same boom-bust pattern that took down BW Resources in 1999. 

ICTSI is a larger, more systemically consequential version of the same mechanism—dressed in blue-chip legitimacy instead of casino-stock notoriety, but running on identical fuel: concentrated, momentum-driven flows mistaken for confidence. 

• Part I called ICTSI the PSEi's single point of vulnerability. 
• Part II showed that vulnerability metastasizing into concentrated liquidity and narrative management.
• Part III shows what happens when the fulcrum that was doing the lifting starts doing the dropping instead. 

The PSEi was supposed to represent the market. But once the representation itself becomes something to be managed—because it communicates confidence and supports financial conditions—it begins to substitute for the reality it was meant to represent. That is the simulacrum. 

The episode reconfirms why the PSEi is becoming an increasingly unreliable gauge of the broader Philippine equity market—and, equally, a more fragile one: susceptible to substantial mispricing, disproportionate liquidity exposure, asset-bubble formation, rising concentration risk, and hidden leverage. 

Benchmark-ism can manufacture confidence for a while. It cannot repeal the arithmetic of the index it's built on.

___

References:

Prudent Investor Newsletters, The PSEi-ICTSI Show, Part II: When One Stock Becomes the Market, July 19,2026 

Prudent Investor Newsletters, PSEi 30: The ICTSI Show, June 7, 2026

 

Stagflation Part 15: The PSEi 30 1H 2026 Earnings Mirage — Nominal Growth, Record Debt, Real Stagnation

 

Because credits springing from fiat inflation provide an easy financial edge, they have the tendency to encourage reckless behavior by the chief executives. This is especially the case with managers of large corporations who have easy access to the capital markets. Their recklessness is often confused with innovativeness—Jörg Guido Hülsmann 

In this issue:

Stagflation Part 15: The PSEi 30 1H 2026 Earnings Mirage — Nominal Growth, Record Debt, Real Stagnation

I. PSEi 30 1H: Record Revenue, Record Debt — And a 2.6% Economy

II. The PSEi 30 divergence

III. Debt: Concentration Makes The Story Worse

IV. Revenue: Where Did The Growth Come From? Meralco’s Money Illusion

V. Price Controls Don't Make The Real Adjustment Disappear

VI. Jollibee: Margin versus Volume Tradeoff

VII. SM Retail's money illusion

VIII. SMC: When Debt Becomes the Growth Mechanism

IX. Q2: Iran War’s Oil Shock Sharpens The Divergence; Three Balance Sheets, One Problem

X. The Corporate Face of Stagflation

XI. Conclusion: The PSEi 30 Earnings Mirage 

Stagflation Part 15: The PSEi 30 1H 2026 Earnings Mirage — Nominal Growth, Record Debt, Real Stagnation 

Record corporate revenue and debt expanded far faster than real GDP growth as policy-suppressed adjustment migrated into prices, margins, investment, and balance sheets. 

I. PSEi 30 1H: Record Revenue, Record Debt — And a 2.6% Economy 

The PSEi 30's 1H 2026 results look remarkably strong — until we ask a more basic question: strong in what sense? 

The figures below are reported corporate results. Even taking the reported numbers at face value, nominal growth in pesos is not, by itself, evidence of real economic growth in output. That is the relevant sense of “money illusion” here: mistaking a change in the unit of account for a change in the underlying quantity—or even quality. Once we make that distinction, the numbers are neither paradoxical nor contradictory. They are consistent with debt-financed nominal expansion occurring alongside weak real growth. 

Revenue surged. Assets surged. Cash surged. Debt exploded! But aggregate earnings fell. And all of this happened while Philippine real GDP grew just 2.6% in 1H and 2.3% in Q2. 

That is not a separate corporate story from the weak GDP, expanding fiscal deficit, and rising public debt already examined in Parts 13 and 14. It is the same story appearing on corporate balance sheets. 

The backdrop is familiar: EO 110's price-suppression scheme, the BSP's five regulatory relief measures, its warnings against NDF speculation, and the defense of the Php 61.75 ‘soft peg’ peso ceiling — all operating while inflation surged and the fiscal and trade deficits widened. Policy can suppress, redistribute, or postpone an adjustment. It cannot repeal the underlying real-resource constraint. 

The PSEi 30 shows where that adjustment appeared on corporate balance sheets. 

II. The PSEi 30 divergence 

1H 2026, aggregate 

Figure 1

Against that: 1H real GDP grew 2.6%, while Q2 real GDP grew just 2.3%. 

Revenue rose more than five times as fast as real GDP. Debt rose more than four times as fast. Net income fell. (Figure 1) 

The corporate sector, therefore, is not experiencing “strong growth” in the real-economy sense. It is experiencing rapid nominal and financial expansion alongside weak real growth. 

And the aggregate numbers conceal an important feature of that expansion: concentration. 

A handful of large conglomerates account for a disproportionate share of the increase in debt, as well as a substantial share of the revenue, assets, and cash behind the index. The result is that the PSEi 30 aggregate is increasingly shaped by the balance sheets of a few very large firms. 

III. Debt: Concentration Makes The Story Worse

Figure 2 

PSEi 30 nonfinancial debt rose by a record Php 654.171B. (Figure 2, upper table and lower chart) 

Twenty of the 26 nonfinancial members added debt. But SMC alone contributed Php 294.1B — roughly 45% of the entire increase — taking its own debt to an unprecedented Php 1.798 trillion, up 19.55% and equivalent to 28.8% of all PSEi 30 outstanding nonfinancial debt. 

The concentration extends beyond SMC. SMC, AEV, and Meralco — the parties to the SMC-AEV-MER gas/power asset transaction, part of the implicit utility bailouts examined last year — together accounted for Php 363.62B, or 56% of the entire increase in PSEi 30 nonfinancial debt. 

Yet, the three largest borrowers overall were SMC, AC, and ICTSI. 

This matters because concentration changes what the aggregate means. 

When the same large balance sheets dominate debt, revenue, assets, and cash simultaneously, the index is no longer a useful proxy for a broad cross-section of independent businesses. A handful of conglomerates increasingly determine the financial appearance of the whole. 

The scale is also significant. The Php 6.251 trillion of PSEi 30 nonfinancial debt is roughly 16.3% of the BSP's total financial-system resources. 

This is not merely a story about leverage inside individual companies. It is a story about the growing weight of a concentrated group of corporate borrowers within the financial system itself. 

SMC is the extreme case. Its Php 1.798 trillion of debt is larger than its entire 2025 revenue and more than eleven times its roughly Php 160 billion equity market capitalization. That is not presented as a conventional measure of debt capacity. It illustrates the mismatch between the financial claims accumulated by the conglomerate and the equity value the market assigns to those claims. 

Acquisitions, asset transfers, refinancing, and debt recycling have expanded SMC's financial structure well beyond the earnings capacity of several of its underlying segments. 

That is where Minsky's framework becomes useful — not as a label, but as a description of the financing process. 

The issue is not simply that SMC carries a large amount of debt. It is that an increasingly large financial structure depends on continued refinancing, asset transactions, and the ability to roll existing obligations forward while earnings growth remains uneven. When operating cash flow is insufficient to service the debt without continued refinancing or the realization of assets, the financing structure evolves to what Minsky called PONZI FINANCE: obligations can no longer be serviced from the cash flows generated by the underlying assets and require new borrowing, asset sales, or other financial transactions to remain current. 

A highly leveraged balance sheet is not automatically a Minsky problem. The problem emerges when the financing structure becomes dependent on the continuation of the financial process that created it. 

The BSP-FSCC's warning about a “wall of maturities” therefore looks different when viewed against this concentration. 

Refinancing risk is not distributed evenly across thirty unrelated companies. A large portion is attached to a relatively small number of very large balance sheets. SMC alone represents an unusually large share of the debt expansion behind the index. 

This also creates a systemic asymmetry. 

When debt becomes concentrated in conglomerates that are economically and politically difficult to allow to fail, leverage can create a form of too-big-to-fail risk even before an actual crisis occurs. The concern is not simply the size of any one company's liabilities. It is the interaction between corporate size, political importance, creditor exposure, and the concentration of those exposures within the banking and financial system. 

The BSP can describe the financial system as “resilient” at the aggregate level while significant fragility accumulates underneath that aggregate. A banking system can remain adequately capitalized while becoming increasingly exposed to the same large counterparties. 

Concentrated corporate leverage therefore exposes concentrated counterparty risk: the failure or forced deleveraging of one major conglomerate can transmit losses through several lenders and financial institutions at once. 

The same concentration also creates a crowding-out problem. 

The issue is not merely that government borrows more. Large corporations and government are drawing on the same underlying pool of financial resources. When conglomerates undertake increasingly large amounts of debt financing without a corresponding increase in productive real investment, they compete with government and other borrowers for savings and bank balance-sheet capacity—coming at the expense of MSMEs. 

That matters because the additional borrowing is not necessarily expanding the economy's productive capacity proportionately. If credit is increasingly being used for acquisitions, refinancing, asset transfers, and balance-sheet restructuring rather than for new productive capacity, the financial claim on future income can grow much faster than the real income available to service it. 

The result is a reinforcing process: 

weak real growth → nominal expansion → heavier corporate borrowing → concentrated leverage → greater refinancing dependence → greater financial fragility. 

The PSEi 30's headline growth therefore becomes less informative the deeper we look into its composition. Revenue is expanding, assets are expanding, cash is expanding, and debt is expanding — but earnings are not keeping pace, real GDP is weak, and an increasing share of the financial expansion is concentrated in a handful of very large borrowers. 

The divergence is not a statistical curiosity. 

It is the balance-sheet expression of the adjustment.         

IV. Revenue: Where Did The Growth Come From? Meralco’s Money Illusion

Figure 3

PSEi 30 revenue surged 13.51% — banks +10.95%, nonfinancials +13.8%. SMC, AEV, and Meralco were the biggest peso revenue gainers. At first glance, this looks like broad corporate resilience. But revenue is a peso measure: it tells us the value of transactions, not the quantity of goods and services behind them. (Figure 3, upper window) 

Meralco makes the distinction almost perfectly. 

Meralco: stagflation in miniature. Physical electricity sales barely moved — Q2 GWh rose just 0.66%, while 1H GWh fell 0.46% — essentially flat to contracting, tracking the broader slowdown. Yet peso electricity sales over the same periods rose 26.4% in Q2 and 17% for the half. (Figure 3, lower graph) 

This is the money illusion in the revenue numbers: the peso value of sales rose dramatically while the underlying physical quantity barely changed. The apparent expansion is therefore much larger in nominal terms than in real activity. 

Income followed the peso line, not the volume line: up 21.14% in Q2 and 12.7% for the half, on revenue growth of 24.65% and 15.7%, respectively. 

The wedge is FIT-All, GEA-All, and other generation/transmission pass-through charges — regulatory and redistributive add-ons, not demand. Consumers pay more through the bill; that additional revenue is redistributed through the system to generators, transmission, and designated energy programs, producing a much higher peso value without a comparable rise in physical consumption. That is the money illusion in concrete form: the peso value of electricity sales rises sharply while the underlying physical quantity barely moves. 

And the "resilient earnings" narrative carries its own balance-sheet cost. Meralco's debt rose 6.5% to a record Php 247.36 billion over the same window. Under the current structure, that increase is not simply additional corporate borrowing in the abstract. It forms part of the balance-sheet transfer associated with the SMC-AEV-MER transaction — the same process examined earlier in the context of the implicit utility bailouts. What appears in the aggregate as stronger revenue and earnings is therefore occurring alongside a redistribution of financial claims and liabilities across the corporate balance sheets. 

Meralco consequently captures both sides of the process in miniature: the nominal value of economic activity rises far faster than its physical volume, while the accompanying balance-sheet transfers create additional financial claims without a comparable expansion of real productive capacity. SMC shows the same process at far greater scale. 

V. Price Controls Don't Make The Real Adjustment Disappear 

Recent BusinessWorld/Bloomberg reporting on companies adapting to tight consumer budgets adds a useful, independent dimension here — corporate behavior confirming the balance-sheet read rather than just corroborating it after the fact. Shakey's Pizza Asia posted a one-third drop in first-half profit and is slowing expansion, citing inflation and fuel costs weakening non-essential spending. Jollibee itself, despite improved customer visits, cut its same-store-sales forecast, profit-growth outlook, and store-opening plans. Monde Nissin is switching to lower-cost ingredients and staggering price increases of 1-5% by product rather than raising prices outright. Century Pacific Food, after holding prices flat for two years, is now raising them 4-5% while planning smaller increases ahead. 

These are companies signaling, through their own operating decisions, that they read the demand environment as weaker than their revenue lines suggest — the same conclusion reached here from the balance sheets, arrived at independently from the boardroom. 

That reporting also points to a broader menu of adjustment worth naming explicitly. When firms can't or won't fully pass higher costs through the sticker price, the adjustment migrates elsewhere: 

• price inflation (raise the price),
• shrinkflation (keep the price, shrink the quantity),
• skimpflation (keep the price, cut the quality — Monde Nissin's ingredient substitution is a live example),
• sneakflation (keep the headline price but raise the effective price through less-visible fees or charges — for example, utility FIT-All and GEA-All charges, or added service, delivery, and platform fees)
• margin compression (absorb the cost),
• cost-cutting (trim inputs, labor, expansion — Shakey's slowing its rollout), or
• balance-sheet expansion (borrow, refinance, or transfer assets to keep the structure moving — SMC and Meralco).

The price can be capped; the loaf can't. It gets smaller, or the ingredients get cheaper, or the margin gets squeezed, or the investment gets postponed, or the debt fills the gap. 

Price suppression can suppress the price adjustment. It cannot suppress the underlying real-resource constraint — the adjustment simply moves to a different line item. 

VI. Jollibee: Margin versus Volume Tradeoff 

Jollibee Food Corporation’s Q2 net income rose 3% to Php 3.519 billion, but 1H net income was still down 16.7% to Php 4.928 billion.

Figure 4

Domestic sales rose 5.61% in Q2 and 6.8% in 1H, despite a 2.7% increase in store count. Against Q2 inflation of 6.8%, real domestic sales were essentially stagnant to negative. More stores generated only roughly inflation-level growth in peso sales. (Figure 4, topmost pane) 

JFC defended profitability where physical volume could not deliver it: margin over volume. That is a rational corporate response to constrained real demand, but it is not evidence of a booming consumer economy. With GDP growing only 2.3% in Q2, the more direct reading is that JFC was protecting margins in an environment where real domestic demand was weak. (Figure 4, middle diagram) 

JFC's balance sheet adds another dimension. Debt rose 13.6% to Php 93.2 billion over the same period. That increase should be read in the context of JFC's entire multinational financial structure — its international expansion and operations, acquisitions, and broader funding requirements — rather than reduced to a purely domestic story. But that broader scope does not make the debt irrelevant to the analysis. It shows that the company's nominal sales and earnings resilience is occurring within a substantially expanding balance sheet.

Jollibee therefore provides another expression of the same process: nominal sales can rise while real domestic volume remains weak; the corporation responds by protecting margins and adjusting operations; and the resulting performance sits within an expanding financial structure that extends beyond the domestic market. The peso-denominated numbers can therefore look resilient without representing comparable growth in the underlying quantity of goods and services. 

VII. SM Retail's money illusion 

SM Retail grew 5.42% in 1H 2026. At the parent-company level, SM Investments' revenue rose 7.53% and net income 6.38%, but those are consolidated figures covering businesses beyond retail. The relevant consumer-sector measure here is SM Retail's own 5.42% growth.(Figure 4, lowest image) 

That number looks less impressive against the inflation environment. Q2 CPI was 6.8% while core CPI 4.1%, meaning SM Retail's nominal growth was below the rate at which consumer prices were rising or slightly above the ex-food and ex-energy CPI. 

In purchasing-power terms, a 5.42% increase in retail activity does not represent real growth if prices were rising faster. 

There is another reason the number deserves scrutiny: SM's retail footprint was expanding at the same time. SM Prime opened two new Philippine malls in 2025 — SM City Laoag and SM City La Union — and opened SM City Zamboanga in March 2026, its 90th Philippine mall. Thus, the 5.42% growth in SM Retail occurred while the broader SM retail platform was adding physical capacity. 

That makes the result a useful indicator of the consumer-side stagflation problem. The business was not merely operating the same stores and selling at higher prices; the broader retail network was expanding as well. Yet SM Retail's nominal growth still lagged Q2 inflation. 

The distinction is the money illusion: more pesos can be recorded as sales without a comparable increase in the quantity of goods purchased. Some of the nominal increase can come from higher prices, while some can come from additional stores and retail capacity. What remains is the underlying real expansion in consumer demand. 

On the available figures, that real expansion appears weak. SM Retail was adding to its physical footprint, yet its nominal growth was still below the prevailing rate of consumer-price inflation. The headline 5.42% therefore overstates the strength of the underlying consumer economy when read without the price effect. 

SM Retail is consequently another expression of the same stagflationary process: the nominal economy grows, but purchasing power and real consumer demand do not keep pace. 

VIII. SMC: When Debt Becomes the Growth Mechanism 

SMC's nonfinancial debt increased by Php 294.1 billion in 1H 2026, reaching Php 1.798 trillion (!). Cash, assets, and revenue all surged. Income did not follow. Of SMC's eight segments, only three posted positive 1H income growth — infrastructure +29%, GSMI +3%, and food +8% — while the rest declined. 

Petron and Global Power delivered substantial revenue growth, but earnings moved in the opposite direction: Petron's fell 27% and Global Power's 7%. Revenue can therefore rise sharply without a corresponding increase in underlying profitability. 

Petron's case is particularly instructive because its revenue growth occurred amid an extraordinary oil-supply shock rather than normal operating conditions. The Iran conflict disrupted traditional Middle Eastern supply routes, prompting Petron to secure alternative sources. It purchased 2.48 million barrels of Russian crude as an emergency measure, with the government encouraging oil companies to find alternative supplies. This was not simply a normal sourcing decision or evidence of a structural improvement in Petron's operating economics; it was part of an exceptional policy and supply response to the disruption in support of EO-110. 

Media accounts that attribute SMC's earnings decline primarily to foreign-exchange losses and one-off gains associated with the SMC-AEV-MER transaction explain why reported income moved during the period. They do not explain the larger divergence: why did SMC add Php 294.1 billion of debt while earnings capacity across the conglomerate remained generally weak?

Figure 5

Nor is the Php 294.1 billion increase a one-off event. SMC's debt has been rising since at least 2013. The largest quarter-on-quarter increase occurred during the first oil shock in 2022, while the Php 129.57 billion increase in Q2 2026 was the second-largest quarterly increase since Q3 2022. The current surge is therefore another stage in an established process of balance-sheet expansion. (Figure 5, upper visual) 

The character of the borrowing matters. The cash-flow pattern shows that SMC is increasingly borrowing to refinance existing obligations, adding to the debt stock while earnings remain weak. This is the mechanism that brings Minsky's Ponzi-finance concept into view: when operating earnings are insufficient to reduce the debt burden, the financial structure becomes dependent on continued refinancing to sustain itself. Debt is no longer merely financing expansion; increasingly, further borrowing is required to maintain the existing financial structure—even as the cost of borrowing rises. (Figure 5, lower chart) 

This also gives the BSP's regulatory-relief measures and its warning about a “wall of maturities” a more concrete significance. The measures can be understood as institutional accommodation of a refinancing problem that has become increasingly important for large, highly leveraged borrowers such as SMC. They provide additional room for maturities to be rolled forward and the adjustment to be deferred; they do not eliminate the underlying liabilities. 

That is the deeper Minskyan concern. Refinancing postpones the adjustment, but does not remove it. When refinancing itself adds to the debt stock, the continuation of the financial structure increasingly depends on the availability of still more financing—and only an easy money environment accommodates this. 

SMC's 1H 2026 results therefore reveal a widening gap between financial (balance sheet) expansion and earnings capacity. Debt, assets, cash, and nominal revenue expanded rapidly while earnings remained generally weak. The balance sheet is no longer simply recording the growth process. Increasingly, it has become part of the mechanism sustaining it. 

IX. Q2: Iran War’s Oil Shock Sharpens The Divergence; Three Balance Sheets, One Problem

Figure 6

Aggregate Q2 net income grew just 1.79% — banks −0.02%, nonbanks +2.27%, with 19 constituents up and 11 down. Revenue, by contrast, surged 18.34% — nonbanks +19.13%, banks +11.99%.  (Figure 6, upper table) 

The Philippine banking system itself grew 7.62% in Q2—propped up by BSP relief measures despite pandemic era financial losses. (Figure 6, lower graph) 

Revenue +18.34% against income +1.79% is the sharpest single number in this piece: nominal activity expanded dramatically while the bottom line barely moved. The divergence is difficult to reconcile with the language of broad-based corporate strength. It is more consistent with an economy in which nominal values and financial claims are expanding faster than the earnings and real activity needed to support them. 

This is where the PSEi 30 closes the loop with the rest of the series. The government expanded its balance sheet to sustain fiscal spending, with public debt growing faster than nominal and real GDP. Banks expanded credit. Corporations expanded debt. Households absorbed record peso consumer loans. Real GDP grew only 2.3% in Q2. 

These are not separate phenomena. They are interconnected balance sheets. Government borrowing creates claims against future fiscal resources. Bank lending creates claims against future household and corporate income. Corporate borrowing creates claims against future corporate cash flows. The financial system can transfer purchasing power across time, finance acquisitions, refinance existing obligations, sustain operating structures, and facilitate asset transfers. But these transactions do not remove the underlying resource constraint; they redistribute claims against it. 

The more important question, therefore, is what happens when institutions repeatedly prevent those claims from being reconciled through structural adjustment? 

That is the significance of the policy sequence examined throughout this series.

• EO 110 suppresses or redistributes price adjustment.
• BSP regulatory relief accommodates stressed balance sheets.
• The exchange-rate regime resists adjustment in the currency.

Each intervention can relieve pressure in the short run, but the underlying imbalance does not disappear simply because its immediate expression has been suppressed. 

The adjustment is displaced — into quantities, quality, margins, investment, debt, refinancing, or other balance-sheet transfers. 

This is the institutional-control problem identified in Luc Lelievre's analysis of why apparently stable systems can become increasingly fragile: interventions that preserve stability at one point in the system can prevent the signals and adjustments through which underlying errors are corrected. 

The consequence is cumulative. When structural adjustment is repeatedly postponed, debt finance increasingly becomes the mechanism through which imbalances are carried forward. Balance sheets expand to absorb pressures that prices, markets, and institutions have been prevented from fully clearing. That can sustain nominal activity for a time, but it also permits malinvestment, misallocation, and financial claims to accumulate against an underlying real economy that is growing much more slowly. 

That is what the H1 2026 data reveal: not simply financial claims growing faster than output, but a system in which policy accommodation is allowing the divergence to persist and increasingly shifting the adjustment onto balance sheets. 

X. The Corporate Face of Stagflation 

The four companies examined above show four different margins of adjustment: 

• Meralco — physical output barely moves while peso revenue surges: price and redistribution. 
• Jollibee — sales barely keep pace with inflation while margins recover: margin over volume. 
• SM Retail — nominal sales growth remains below inflation despite an expanding retail footprint: money illusion and weak real demand. 
• SMC — debt, assets, cash, and revenue expand while earnings remain generally weak: balance-sheet expansion and refinancing. 

And beneath all four sits an adjustment that does not necessarily appear in headline financial statements: shrinkflation, skimpflation, sneakflation, margin compression, deferred investment, lower-quality substitution, and debt accumulation. The loaf gets smaller; the ingredient gets cheaper; the product becomes thinner; the expansion is postponed; the fee appears somewhere else; or the balance sheet absorbs the pressure. 

The price can be suppressed. The adjustment cannot. 

That is why the corporate response to stagflation cannot be read from CPI alone. The adjustment can move through price, quantity, quality, margins, investment, or debt, and different companies are choosing different combinations of the same underlying menu. 

What looks like stability in one line of the accounts can therefore represent adjustment somewhere else. And when policy repeatedly suppresses the visible adjustment, the hidden adjustment increasingly migrates into balance sheets — where it appears first as accommodation, then as leverage, and eventually as fragility. 

XI. Conclusion: The PSEi 30 Earnings Mirage 

Strong in what sense? 

The PSEi 30's 1H 2026 results look robust: revenue reached Php 3.974 trillion, assets Php 33.417 trillion and cash Php 1.608 trillion, but nonfinancial debt Php 6.251 trillion. 

Yet revenue grew 13.51% while real GDP grew only 2.6% in 1H and 2.3% in Q2. Debt grew 11.69% while aggregate net income fell 1.29%. The headline financial expansion is therefore not evidence of equivalent real economic expansion. 

That is the money illusion at the center of the PSEi 30. Nominal values rise and create the appearance of growth even when the underlying quantity or quality of economic activity does not rise proportionately. Meralco's peso electricity sales rose sharply while physical consumption barely moved. Jollibee's domestic sales grew roughly at the rate of inflation. SM Retail's nominal growth remained below inflation despite an expanding retail footprint. SMC's debt, assets, cash, and revenue expanded while earnings remained generally weak. 

But the money illusion does not arise in isolation. It is the financial appearance produced by the stagflationary process. Real growth remains weak while prices rise, purchasing power is constrained, and the adjustment that would ordinarily expose the imbalance is repeatedly suppressed, redistributed, or postponed. EO 110, BSP regulatory relief, and the exchange-rate regime operate on different parts of that adjustment process. They can alter where the pressure appears without eliminating the underlying constraint. 

The result is a displacement of adjustment. It appears in prices, volumes, product quality, margins, investment, debt, and refinancing. What cannot be absorbed through the price is absorbed through quantity; what cannot be absorbed through quantity is absorbed through quality or margins; what cannot be absorbed operationally can migrate onto the balance sheet. Financial expansion can therefore continue even while real economic expansion remains weak. 

That is what the PSEi 30 adds to the stagflation series. The corporate sector does not merely reflect weak growth and inflation; its balance sheets show how the economy is absorbing the adjustment. Nominal revenue can surge while real demand stagnates. Debt can expand while earnings weaken. Financial claims can accumulate while productive capacity and real output lag behind. 

The price can be suppressed. The adjustment cannot. 

The PSEi 30 is where that adjustment becomes visible in corporate form: stagflation underneath, money illusion on the surface, and balance-sheet expansion in between. 

____

References: 

Prudent Investor, Stagflation Part 14: Q2 GDP at 2.3% and Falling — A GDP Built on Suppression and Debt, August 9, 2026 

Prudent Investor, Stagflation Part 13: Record Twin Deficits Signal the Cost of Deferred Adjustment, August 2, 2026 

Prudent Investor, Inside the SMC–Meralco–AEV Energy Deal: Asset Transfers That Mask a Systemic Fragility Loop, November 23, 2025 

Prudent Investor, The Oligarchic Bailout Everyone Missed: How the Energy Fragility Now Threatens the Philippine Peso and the Economy, December 7, 2025 

Luc Lelièvre, Why Stable Systems Fail: The Illusion of Institutional Control, Mises.org, May 18, 2026

 

Why Isn't Gold Acting Like a Safe Haven—Yet? Part III: Is This Gold's Next Leg Up?

 

First of all there is need to remember that the gold standard did not collapse. Governments abolished it in order to pave the way for inflation. The whole grim apparatus of oppression and coercion—policemen, customs guards, penal courts, prisons, in some countries even executioners—had to be put into action in order to destroy the gold standard. Solemn pledges were broken, retroactive laws were promulgated, provisions of constitutions and bills of rights were openly defied. And hosts of servile writers praised what the governments had done and hailed the dawn of the fiat-money millennium.—Ludwig von Mises 

In this issue: 

Why Isn't Gold Acting Like a Safe Haven—Yet? Part III: Is This Gold's Next Leg Up?

I. Introduction: From Finding the Bottom to Explaining the Reversal

II. The Gold Cycle: Cyclical bear, Secular bull

III. The New Plaza Accord: Intervention and Its Shrinking Half-Life

IV. Deepening Interventions Becomes Gold's Catalyst

V. When Former Headwinds Become Tailwinds

VII. Central Banks as Dip Buyers, and the Fund-Flow Tailwind

VI. Risk-Off, Inflationary, and Fiscal Signals Converge

VII. Central Banks as Dip Buyers, and the Fund-Flow Tailwind

VIII. Bottom line: Gold Is Pricing the Limits of Intervention 

Why Isn't Gold Acting Like a Safe Haven—Yet? Part III: Is This Gold's Next Leg Up? 

Intervention was supposed to stabilize currencies and bonds. Instead, each attempt is exposing the imbalance underneath—and gold is beginning to price it.

I. Introduction: From Finding the Bottom to Explaining the Reversal 

Figure 1 

This piece started as an attempt to answer whether gold had found a bottom. Events overtook the question. In the space of a week, gold cleared its 200-day moving average at $4,501, broke above $4,600, surpassed the 38.2% retracement of its decline from March's record, and confirmed the rounding-bottom pattern that had been forming since June. It was gold's third straight weekly gain—and its third strongest week of the year. 

The question of the bottom is no longer the interesting one. 

What's worth asking now is whether this is the start of gold's next leg toward its late January record near $5,400-$5,600, and — more importantly for this series — why it's happening.

Part I and Part II of this series argued gold's early-2026 weakness was a liquidity-stress phenomenon, not a failure of its safe-haven role — the gold-oil ratio compressing as energy-importing economies scrambled for dollars, gold sold because it remained liquid enough to fund margin calls, not because confidence in it had broken. Gold peaked above $5,500 intraday in late January, fell in a rounding top that hardened into a roughly 28% decline into late June, then based and turned. That much was covered in the earlier articles. 

What has changed is the pace and the trend—and, as the following analysis shows, the trigger. 

II. The Gold Cycle: Cyclical bear, Secular bull 

Gold's advance since the Nixon shock of August 15, 1971, has never moved in a straight line. It has progressed through a series of secular cycles, with cyclical bear markets occurring within the larger structural trend. Those corrections have varied considerably in both depth and duration.

Figure 2

The shortest cyclical bear market under a secular bullmarket was the 2008 decline, at roughly 26%, which resolved within about eight months. The deepest was the 2011–2015 decline, at roughly 48%, which lasted four years and marked the hiatus before the current cycle. February–June's roughly 28% decline over four months therefore looks much closer, in depth and duration, to the 2008 cyclical correction than to the 2011–2015 secular break. (Figure 2, topmost pane) 

Importantly, the structural forces behind the current advance have not disappeared. Central-bank de-dollarization, fiscal dominance, a geopolitical order drifting toward blocs, a deepening war economy, asset bubbles, and persistent inflation remain part of the monetary and institutional environment supporting gold. 

A liquidity-driven correction can therefore interrupt a secular advance without terminating the forces that produced it. If the current reversal holds, the February–June decline may prove to have been another cyclical bear market nested within the larger secular bull—not the end of the cycle itself. 

And if that interpretation is correct, the current advance would represent the third major leg of gold's post-Bretton Woods secular bull, following the 1970s advance and the 2001–2011 bull market. (Figure 2, middle graph) 

III. The New Plaza Accord: Intervention and Its Shrinking Half-Life 

On July 31, Japan's Ministry of Finance confirmed a coordinated yen-buying intervention with the US—the first joint action since 2011—after the yen slid toward 40-year lows near 163 to the dollar. 

The September 1985 Plaza Accord is the obvious comparison, but the history deserves care. Gold began rising in February 1985, before the Accord, and rallied roughly 50% from the Accord through November 1987—so gold did respond. (Figure 2, lowest image) 

What the Accord did not do was deliver gold's secular low. The dollar devaluation it engineered helped inflate Japan's property and stock bubble, which peaked in 1990 and produced the “lost decades,” with aftershocks plausibly extending through the 1994 Tequila crisis, the 1994 bond crisis (Great Bond Massacre), and the 1997 Asian financial crisis. 

Gold's actual secular bottom wasn't etched until the post-Asian-crisis, post-dot-com years around 1999–2001, symbolically marked by Gordon Brown's UK gold sales near the trough. 

The 1980–1999 gold secular bear was, in effect, the salad days of the dollar standard. 

A Plaza-style intervention can produce a real, tradeable cyclical gold rally without stabilizing the underlying system. Plaza's deeper legacy was several years of instability surfacing elsewhere, not durable currency-regime repair.

Figure 3

The yen's own history since 2022 makes the point more directly. Japanese authorities have repeatedly intervened to support the yen, yet each intervention has ultimately been followed by renewed downward pressure. (Figure 3, topmost pane) 

The pattern has been remarkably consistent: intervention produces an abrupt reversal, but the underlying trend eventually reasserts itself. 

That is the shrinking half-life in practice. The more fundamental forces driving the currency remain in place, so each intervention has to work harder to produce the same effect—and the market increasingly treats the intervention itself as evidence of the underlying imbalance. 

That caution has aged well. It took the bond market almost two weeks to fully erase the initial effect of the Fed-BoJ intervention. Then, on August 19, with 30-year Treasury yields pushing above 5.30%—the highest since 2007—the Treasury unexpectedly doubled its long-end buyback capacity, from a $2 billion to at least a $4 billion cap per operation, effective September 9. Yields fell instantly, equity futures jumped, and gold spiked. 

The Treasury's buyback program had already been operating since May 2024. The August move therefore wasn't the introduction of a new tool, but an expansion of an existing one. The important point is what happened despite that intervention: long-term yields had continued rising, suggesting diminishing returns from efforts to absorb duration and stabilize the long end. (Figure 3, middle image) 

This time the effect lasted one day! By Friday, the 30-year yield had regained the entire drop and then some, closing at 5.27%, while the 10-year sat at 4.74%—a single basis point from its pre-intervention high. 

Wolf Richter's account of the episode is worth noting for its diagnosis: the rise in long-term yields wasn't market dysfunction requiring correction, but the consequence of the government having to sell roughly $1 trillion in new debt over three months to fund the deficit, with buyers demanding compensation for inflation risk, the fiscal trajectory, and the sheer volume of paper that must clear the market every few months. Those pressures cannot be resolved by a signaling exercise from the Treasury secretary. 

Treasury Secretary Bessent himself all but conceded the point on CNBC the following morning, repeatedly describing the move as "signaling." 

Nor is Treasury the only official-sector actor absorbing government debt. The Federal Reserve's balance-sheet reduction ended in December 2025, after which it began purchasing Treasury bills and other short-term Treasuries to maintain an ample supply of reserves. Its outright Treasury holdings have subsequently been rising, reaching roughly $4.54 trillion by August 19, 2026. FRED (Figure 3, lowest visual) 

This may not be technically classified as quantitative easing (QE), but economically it represents renewed official-sector absorption of Treasury securities. And the more revealing point is what has happened alongside it: long-term Treasury yields have continued to rise. 

The implication is straightforward: despite the labeling, official-sector absorption of Treasury debt is increasing, yet it is not preventing the market from demanding higher yields on longer maturities. That is evidence of diminishing intervention effectiveness, not evidence that the underlying pressure has been resolved.

Figure 4 

And the underlying pressure is not static. The US national debt has now crossed $40 trillion, while interest costs have continued to accelerate. According to the figures cited by ZeroHedge, interest expense had already reached roughly $1.37 trillion in 2026, about 20% above the comparable period a year earlier. (Figure 4, topmost and middle images) 

This is where intervention encounters a structural problem. Deficit spending creates the borrowing requirement; rising yields increase the cost of servicing that borrowing; and higher interest costs, in turn, enlarge the deficit and require still more borrowing. The market is therefore confronting a moving target: official-sector intervention may temporarily absorb Treasury supply or suppress yields, but the fiscal requirement generating that supply continues to expand. 

The effectiveness of intervention consequently diminishes as the underlying imbalance grows. What might once have been sufficient to stabilize the market becomes progressively less effective when the volume of debt, the interest burden, and the compensation demanded by investors are all rising together. 

Two interventions in three weeks have produced only temporary relief, while the fiscal, currency, and bond-market pressures behind them have quickly reasserted themselves. That is the more important signal for gold. Yields of 10 and 30 year treasuries were at milestone highs (Figure 4, lowest diagram)

Figure 5 

Japanese Government Bons (JGB) yields tell a similar story. 

They remain below their immediate pre-intervention spike, but have begun climbing again, touching a fresh three-decade high near 2.94% on August 18 before easing only slightly. Figure 5, topmost image) 

The same pattern is emerging in Japan: intervention can alter market prices temporarily, but it does not remove the underlying pressure on the currency and sovereign bond market. 

IV. Deepening Interventions Becomes Gold's Catalyst 

For gold, however, the significance is different. 

Gold had already been trading around and holding the $4,000 level from late June through early August. 

First, the Fed-BoJ intervention provided the catalyst for the first lift-off from that base, reinforcing $4,000 as the floor. 

Then, Bessent's Treasury backstop came afterward, adding fuel to an advance that was already underway. 

The initial decline in long-term yields quickly disappeared, but gold retained its momentum. 

The result was the sequence described at the beginning of this piece: gold broke its 200-day moving average at $4,501, moved above $4,600, surpassed the 38.2% retracement of its March decline, and confirmed the rounding-bottom pattern that had been forming since June. 

The significance therefore goes beyond the failure of either intervention to suppress yields. The interventions themselves are becoming part of the mechanism driving gold higher. Attempts to contain currency and bond-market pressures are adding to the monetary and financial imbalances that gold is increasingly pricing. 

This is where the comparison with the Plaza Accord becomes more interesting. The 1985 agreement did not simply weaken the dollar; it set in motion adjustments that eventually surfaced elsewhere in the financial system. 

Today's bilateral intervention comes against a far more leveraged fiscal and monetary backdrop, with sovereign debt, currency instability, and official-sector intervention increasingly interacting with one another. 

If history rhymes, the contemporary bilateral Plaza Accord may mean more than a break of the recent high—it may mark the beginning of sharply higher gold prices. 

V. When Former Headwinds Become Tailwinds 

WTI crude was up roughly 5% this week, back above $86, as the US-Iran standoff over the Strait of Hormuz shows no sign of resolving and Washington prepares tighter sanctions. (Figure 5, middle graph) 

That is worth pausing on because oil-driven dollar demand was the mechanism Part I and Part II identified as suppressing gold earlier this year: energy importers scrambled for dollars, and gold was sold for liquidity even as the safe-haven case strengthened. 

The same forces are now coinciding with gold's advance rather than working against it. 

That is not a contradiction. It means the liquidity-stress channel has been overtaken by the larger fiscal-currency-risk channel this piece has been tracking. The same energy shock that once forced gold sales for dollar liquidity is now compounding, rather than competing with, the debt and currency concerns driving the bid for gold. 

There is another important change in the relationship between gold and Treasury yields. Earlier in the year, rising 30-year Treasury yields appeared to place a ceiling on gold's advance. Now the relationship appears to be changing: gold is rising even as the 30-year yield pushes above the level at which the Treasury's intervention sought to stabilize it. 

The market is therefore no longer responding to higher yields simply as an opportunity cost for holding gold. It is increasingly interpreting those yields as evidence of the fiscal and currency pressures that make gold more attractive in the first place. 

It is also worth noting that rising gold prices and rising 10-year Treasury yields have historically coexisted during the stagflationary periods of the 1970s. (Figure 5, lowest chart) 

The parallel with today is not exact—the degree of systemic leverage is vastly different—but the coexistence of rising nominal yields and rising gold is not unprecedented when inflation, fiscal pressure, and confidence in the monetary regime become dominant forces. 

VI. Risk-Off, Inflationary, and Fiscal Signals Converge 

Gold's rise also came as risk appetite weakened. The S&P 500 fell about 1.4% on the week, its first weekly decline in a month, dragged by a rough five days for technology; the Nasdaq lost roughly 2%. That gives this week's move a stronger risk-off component: equities fell while gold rose. 

But it was not a conventional flight to safety. Oil was rising, long-term Treasury yields were rising, and gold was rising with them. Growth concerns and inflation/fiscal concerns were therefore appearing simultaneously: stocks were pricing weaker risk appetite, while oil and bond yields were pricing inflation, supply, and fiscal pressures. 

That combination is arguably more important for gold than a conventional risk-off episode. Gold was not merely benefiting from falling risk assets; it was responding to the same underlying deterioration that was simultaneously pushing up oil and long-term yields. 

Curiously, Bitcoin also surged more than 22% during the week, briefly moving above $77,000, reflecting both renewed risk appetite in the crypto market and, potentially, a safe-haven bid amid concerns over currencies and the monetary system. Short covering amplified the move. 

Taken together, the week's price action looks less like a simple flight to safety than a convergence of risk-off, inflationary, and fiscal signals—and gold is increasingly responding to all three. 

VII. Central Banks as Dip Buyers, and the Fund-Flow Tailwind

Figure 6

Central banks were buying into the selloff. The World Gold Council's Q2 2026 data showed central banks adding 102 tonnes of gold during the quarter, even as gold prices fell roughly 16%. (Figure 6, upper diagram) 

As of the first half of 2026, Poland remained the largest reported buyer at 82 tonnes, followed by Uzbekistan at 41 tonnes, China at 40 tonnes, and Kazakhstan at 27 tonnes. World Gold Council: Central Bank Gold Statistics, June 2026 (Figure 6, lower graph) 

The significance is not the absolute volume, but the direction of demand during the correction. While more price-sensitive and leveraged holders were liquidating, official-sector demand was increasing. Central banks were therefore absorbing part of the supply released by the selloff, providing an important counterweight to the liquidation pressure that drove gold lower. 

The longer-term demand signal remains equally important. The World Gold Council's survey found that 89% of reserve managers expect official gold holdings to increase over the next 12 months. That suggests the buying is not simply a reaction to short-term price movements, but part of a broader shift in reserve preferences.

Figure 7

Fund flows provide a second layer of support. Gold held the $4,000 area for close to two months as renewed ETF inflows emerged, while the two previous stretches of stronger fund flows—roughly August–October 2025 and November–February 2026—preceded significant advances, including the run into January's record. (Figure 7, upper image) 

The pattern does not mean ETF flows mechanically determine prices, but it does show that renewed investment demand has tended to accompany—and at times precede—the next major leg higher.  

Independent ETF data reinforce the change in direction: global gold ETFs returned to inflows in July after May–June redemptions had flushed out leveraged positions, with flows continuing into August. The important development is therefore not simply the amount of money flowing into gold, but the transition from forced liquidation to renewed accumulation. 

India's Dhanteras and Diwali season could provide an additional, though more modest, seasonal catalyst into year-end, alongside whatever short covering remains from a positioning backdrop that has been reduced but does not appear fully washed out. (Figure 7, lower visual) 

The ownership dynamics are therefore changing. Central banks were accumulating while weaker and more leveraged holders were liquidating; as that selling pressure subsided, ETF flows turned positive. The correction consequently did not destroy the underlying demand structure. Instead, it transferred gold from more price-sensitive holders toward buyers with a stronger strategic (non-price sensitive) incentive to own it. 

That matters for the next leg. The same selloff that removed speculative excess also created the conditions for a stronger base: official buyers absorbed supply, leveraged positions were reduced, and investment flows are now returning as gold moves higher. 

VIII. Bottom line: Gold Is Pricing the Limits of Intervention 

The question this piece opened with has effectively been answered by the market: gold has bottomed. What deserves scrutiny now is why, because the explanation is not the conventional one. 

Two interventions in three weeks—one in currencies, one in the bond market—have produced only temporary effects in the markets they were intended to stabilize. Yet their significance for gold has been the opposite. 

The Fed-BoJ intervention provided the catalyst for gold's first lift-off from the $4,000 base; Bessent's Treasury backstop then added fuel to an advance already underway. The interventions did not extinguish the underlying pressures. They exposed them—and, in doing so, reinforced the forces driving gold higher. 

The same energy shock that once forced gold into liquidity sales is now moving with gold rather than against it. This week's equity sell-off adds another layer: stocks fell while gold, oil, and long-term yields rose together. That is not a simple flight-to-safety trade. It is a combination of risk-off, inflationary, and fiscal signals—a more distinctly stagflationary configuration. 

The next leg will not necessarily be linear. $4,600 now needs to establish itself as support rather than prove to have been a temporary spike, while a sustained break below $4,300 would materially weaken the bullish structure. But the more important question is no longer technical. 

Gold is increasingly pricing the widening gap between what governments can credibly promise about their currencies and debt markets and what they can actually deliver. The interventions are themselves becoming part of that price signal: each attempt to contain currency or bond-market pressure produces a temporary market response, while the underlying imbalance quickly reasserts itself. 

And that is ultimately what makes this leg different from the earlier stages of the cycle. Gold is no longer simply rising because investors fear what might happen. It is rising because the market is beginning to price what governments are already doing—and the diminishing effectiveness of their attempts to contain the consequences. 

____

References

-Why Isn’t Gold Acting Like a Safe Haven—Yet? The Gold–Oil Ratio and the Liquidity Stress Behind Early-Crisis Gold Weakness (Part II)—April 5, 2026 

-Why Isn’t Gold Acting Like a Safe Haven—Yet? War, Liquidity Stress, and the Fracturing of the Bullion System (Part I)— March 22, 2026