Jackson Hole and the 16 September hike: the Fed follows the market more than it leads it
Three weeks after a speech in which its new Chair put the fight against inflation first, the US Federal Reserve raised its policy rate for the first time since July 2023. We see this less as a return to monetary orthodoxy than as the central bank catching up with the market: investors had anticipated the decision, the constraint of public debt remains intact, and gold has held up well even though the risk we identified in our August article materialised.
For private wealth investors and allocation professionals
KEY POINTS
• The US Federal Reserve raised its policy rate on 16 September, but markets had already anticipated the move: we see a central bank catching up with the market rather than a deliberate tightening of policy.
• In our view, the burden of public debt limits the Fed’s room for manoeuvre: every rate rise adds to the government’s interest bill, just as inflation is picking up again.
• Gold has stabilised around its mid-August level. We still expect gold to move back above USD 5,000 an ounce, but our timing has shifted into 2027.
• Shares in gold mining companies are attracting investors again, yet for the same expected earnings they cost nearly three times less than technology shares.
• These shares would offer no protection in the first phase of a credit crisis, when investors sell whatever they can: the choice of companies, based in particular on their level of debt, then becomes decisive.
WHAT HAPPENED
On 28 August, in his first address to the Jackson Hole symposium as Chair of the Federal Reserve, Kevin Warsh distanced himself from a practice established by his predecessors, that of announcing in advance the likely path of policy rates. Such guidance, in his view, has its place in a crisis but risks misleading markets in normal times. He also set out two principles to which we return below. The first is that the 2% inflation target, measured by the personal consumption expenditures (PCE) price index, is not negotiable. The second is that the growth of the money supply should once again be an indicator the central bank follows closely, an idea he himself acknowledged has fallen out of fashion.
On 16 September, by a unanimous vote of its twelve voting members, the Federal Open Market Committee raised the target range for its policy rate by a quarter of a percentage point, to 3.75-4.00%. It was the first increase since July 2023. According to their own projections, most committee members expect a further increase by December, followed by stable rates next year.
Gold fell in response to the announcement, then stabilised around its mid-August level. Having closed at USD 4,657 an ounce on 25 August, it fell to a close of USD 4,264 on 16 September, rose to USD 4,379 on 18 September before easing back to USD 4,299 on 22 September, against USD 4,376 in mid-August. The scenario our previous article presented as the main risk has therefore materialised, without triggering the renewed decline we feared.
A CENTRAL BANK CATCHING UP WITH THE MARKET
Many saw this rate rise as the first sign of a Fed returning to orthodoxy. We do not share that reading, simply because of the timing: the market had anticipated the move. On 15 September, the yield on two-year US government bonds, which reflects the average level of policy rates investors expect over that period, already stood at 4.66%, roughly one percentage point above the effective federal funds rate, the policy rate actually in force, which was then 3.63%. In raising rates, the central bank was therefore not leading the market but falling into line with it. Conversely, leaving rates unchanged would probably have been read as an inadequate response to inflation, pushing long-term rates higher at the very moment when the cost of public debt makes such an increase particularly damaging. Nor did the gap close after the decision: on 18 September, the two-year yield stood at 4.74%, against an effective rate that had risen to 3.88%. In other words, the market still expects higher policy rates than those the Fed has so far decided.
The Jackson Hole speech supports this reading. Kevin Warsh acknowledged that he would find it hard to describe financing conditions in the US economy as restrictive: the gap between the rates paid by companies and those paid by the government is close to its historical lows, and banks are lending to businesses on terms among the most generous ever observed. He also noted that inflation as measured by the PCE index, the gauge the central bank targets, stands at 3.7% over twelve months but at 4.1% annualised over the past six months, meaning that it is accelerating again, and that it has exceeded the Fed’s 2% target for sixty-five months, more than five years. He added that the recent rise in commodity prices warrants attention. Brent crude, still below USD 90 a barrel at the time of the speech, went on to rise above USD 100 in the week before the September meeting. Finally, he pointed out that business investment in equipment and in intangible assets such as software is growing by around 9% a year, with more than half of that growth attributable, in his view, to artificial intelligence. That investment has to be financed, just as the federal government has to finance its own large deficit. In our view, companies and the government are therefore competing for the same pool of available capital, which puts upward pressure on rates over and above the effect of the Fed’s own decisions.
The monetary data tell the same story. Annual growth in the amount of money in circulation in the United States, as measured by the M2 aggregate, which covers notes and coins, deposits and readily available household and corporate savings, rose to 5.4% in July, from 4.4% a year earlier and 4.5% as recently as April. It remains below its average since 1960, of around 6.8%, but is closing in on it quickly. The money supply is thus accelerating at the very moment the Fed Chair says it should be watched, a sign that monetary conditions remain loose. Against that backdrop, a quarter-point increase amounts to limited tightening rather than a real turn of the screw.
This modest increase also worked to the Fed’s advantage. A newly appointed Chair, whose independence is under close scrutiny, could hardly begin his term by leaving rates unchanged, a choice that would have fuelled doubts about that independence. We see the September move as the start of a shift in the Fed’s message: the central bank is presenting itself as a bulwark against inflation. The change of tone came all the more easily because the market had already anticipated the increase. This is our interpretation, not an intention stated by the central bank. The new stance will nonetheless be judged by the decisions that follow. Yet those decisions remain subject to the same constraints, starting with the weight of public debt.
THE INFLATION TARGET AND HOW IT IS MEASURED
One of the first decisions to test this new stance could concern the very way inflation is measured. At his Senate hearing before his appointment, Kevin Warsh spoke in favour of so-called trimmed means, which each month set aside the most extreme price increases and decreases in order to capture the underlying trend more accurately. The gap between such a measure and the main indices in use today is far from negligible. In July, the trimmed mean calculated by the Federal Reserve Bank of Dallas stood at 2.3% over twelve months, while the PCE index excluding food and energy rose by 3.3% and the headline index by 3.7%.
We are not suggesting that anyone intends to bring inflation down by changing the way it is calculated. Indeed, at Jackson Hole the Chair drew on a detailed analysis of prices to paint a harsher picture: 54% of the goods and services in the basket are rising by more than 3% a year, compared with an average of 32% in the two decades before the pandemic. The fact remains that the 2% target has just been reaffirmed by a Chair who, in assessing the underlying trend, favours a measure that currently puts inflation 1.0 to 1.4 percentage points below the headline and core PCE indices. Were such a measure to carry more weight in the committee’s decisions, measured inflation would appear much closer to the 2% target, without prices having slowed at all. The central bank might then judge that less tightening would be enough to bring measured inflation back to the 2% target, which would considerably widen its room for manoeuvre.
THE WEIGHT OF PUBLIC DEBT
If the central bank is being so cautious, it is first and foremost because of debt. The interest the federal government pays each year on its borrowing now exceeds USD 1 trillion, so that every rate increase adds to the budget bill and monetary policy is gradually becoming dependent on the management of that debt. The US Treasury provided an illustration of this over the summer. On 19 August, it announced that it would at least double the maximum size of its buyback operations in ten- to thirty-year government bonds. By buying back these bonds, the Treasury supports their price, which lowers their yield, in other words the long-term interest rate. But the fall in long-term rates that followed the announcement lasted a single session. The Fed is thus raising short-term rates while the Treasury tries to bring long-term rates down, so that monetary policy and Treasury policy are working against each other. In our view, this is the most concrete manifestation of what is known as fiscal dominance, the situation in which the government’s financing needs end up dictating monetary policy.
At the same time, the composition of central bank reserves has changed. According to the European Central Bank, gold accounted for 27% of official reserves at the end of 2025, against 22% for US government bonds, which had not been the case since the mid-1990s. The shift owes most to the rise in the gold price, which mechanically revalues existing holdings: at end-2023 prices, US government bonds would still be well ahead, at 26% against 16% for gold. Gold nonetheless occupies a growing place in global reserves and is regaining, in practice if not formally, its role as a monetary asset.
OUR AUGUST EXPECTATION
In our article of 17 August, we expected gold to return above USD 5,000 before the end of the year, on condition that the Fed neither raised rates in September nor signalled such a move at Jackson Hole. We also identified a rate increase as the main risk of a renewed decline in gold. The rate increase duly came, but the decline in gold did not. As the condition has not been met, we can no longer hold to our year-end timeline with the same confidence, and we prefer to say so plainly. The price level we expect, however, is unchanged. We noted in August that UBS was targeting a comparable level for the first half of 2027 and that our disagreement concerned the timing, not the level. The 16 September decision now leads us to envisage a timeline closer to that of UBS, with no change to the level we expect.
MINING EQUITIES
The most telling signal in favour of our scenario comes from mining shares. On 25 August, Newmont, the world’s largest gold producer, reached an all-time intraday high of USD 135.29. On the same day, gold posted its summer high, which was still some USD 760 below the closing high it had reached in January. When a company sets a new record while the metal from which it earns its revenue remains well below its own record high, the most plausible explanation is an inflow of capital into the sector. Investors are returning to the sector, starting with the largest names: they are anticipating further gains rather than reacting to the current gold price. Newmont is not an isolated case, since the ratio of the gold miners’ index to the gold price also peaked on 25 August, at its highest level since August 2016, which means that gold mining shares had risen faster than the metal itself.
These flows are no accident. Newmont reported record free cash flow of USD 2.2 billion for the second quarter and has reduced its share count by around 9% since February 2024. Gold miners nonetheless remain far cheaper than the rest of the market: their enterprise value stands at only 6.0 times expected earnings before interest, taxes, depreciation and amortisation (EBITDA), against 17.4 times for the global technology sector. In other words, for the same level of earnings, investors are paying nearly three times less for gold miners. Equity fund managers who invest across the whole market have stayed away from the sector for the past ten years. If they join the investors who have already repositioned in the large gold producers, the market for gold mining shares is too narrow to absorb the new capital without prices rising. Such revaluations tend to be abrupt rather than gradual. As for silver, it generally follows the same path as gold when gold is rising, but with larger swings.
AN OBJECTION TO TAKE SERIOUSLY
This favourable view of gold mining shares nonetheless faces a serious objection. If monetary tightening continues, as the Fed’s projections suggest, it will eventually weigh on a heavily indebted economy. A credit crisis, should one follow, would then be very different from an ordinary stock market correction. Some fear such a crisis in the private credit market in particular, where funds lend directly to companies without going through banks. In the first phase of such a crisis, investors who have to put up additional collateral to cover their commitments sell whatever they can, gold included. We recalled in August that the gold miners’ index held up better than the S&P 500 both in 2008 and in March 2020, falling by 31% and 26% respectively, against 57% and 34% for the US index. Over these two episodes as a whole, however, gold itself did far better, gaining 25% during the first and losing only 4% during the second. An investor who expected mining shares to provide immediate protection against such a crisis would therefore be disappointed.
What then sets companies apart is their ability to come through this first phase without having to raise new capital. On this point, the financial position of the large producers differs from that seen in earlier periods of stress. They now carry little debt relative to the cash they generate, and their capital spending goes into extending existing mines rather than financing new projects. They therefore depend less on credit than in the past, when a drying up of financing forced them to issue shares at the worst possible moment. Companies that have yet to build their mines are in a different position: by definition they depend on external financing, so they are the ones that would suffer most from a rise in the cost of credit. In normal times, the distinction between producers already in operation and these development-stage companies is of secondary importance. In such a crisis, it would become decisive. It is one of the criteria that guide the selection of companies in the funds we manage. We favour producers already in operation, particularly those with expansion projects due to come on stream within two to three years. Among companies not yet in production, we mainly select those close to production rather than exploration companies, which are still searching for deposits. As a rule, we therefore do not select exploration companies, however promising, when production is still too far off: we do not invest solely on the assumption that a major group will eventually acquire them at a premium to their share price.
What happens after that initial phase of forced selling is harder to predict. We observed in August that, in the episodes of stress mentioned above, mining shares had behaved far less consistently after their low point than during the fall. Because those episodes showed no consistent pattern, we drew no conclusion from them, and we draw none today. If we expect gold mining shares to outperform gold itself once that phase has passed, it is because of a mechanism rather than a historical regularity. On the one hand, a rise in the gold price lifts a producer’s earnings by a larger proportion than the rise in the metal price itself, because its revenue follows the metal price while most of its costs do not rise at the same pace. On the other, a central bank response to a credit crisis would recreate precisely the conditions that support the gold price, namely falling real interest rates, that is to say rates adjusted for inflation, which make interest-bearing investments less attractive relative to gold, since gold pays no interest, and more abundant liquidity, which supports demand for the metal. Silver tends to amplify movements in the gold price, both up and down, with one important difference: much of its demand is industrial and therefore tied to the economic cycle, which would also make it the metal most exposed to a recession.
Finally, should our scenario be slow to materialise, the valuation gap would remain an argument in the sector’s favour. Mining companies whose net margins exceed those of the broader market, 28.8% against 13.7% for the S&P 500 as at 22 September, but which trade on considerably lower valuation multiples, do not need extreme assumptions to justify their place in a private wealth allocation. That is what makes the wait bearable.
THE SIGNS WE WILL BE WATCHING
We do not know when the constraint imposed by public debt will prove stronger than the central bank’s message, and we therefore refrain from putting a date on it. We nonetheless judge that moment to be closer than current market pricing implies. Although we cannot date it, we do know the signs by which we will recognise it. The first would be a repeated failure of the Treasury’s buybacks to bring long-term rates down on a lasting basis. The second would be a widening of the gap between the rates paid by companies and those paid by the government. Its current low level shows that lenders are not yet demanding a higher premium to lend to companies, which is why the economy has so far absorbed higher rates without major strain on credit. The third would be the two-year yield falling below the policy rate. Because that yield reflects expected policy rates, such a move would show that investors were anticipating rate cuts, and therefore that they judged tightening to have gone too far. When the Treasury’s debt management is no longer enough to contain long-term rates, the authorities are likely to have to intervene more directly, by capping long-term rates or by creating money to buy back public debt. At that point, the case for gold will no longer rest on conviction alone: capping rates or creating money would bring about precisely the conditions that support its price.
WHAT WE TAKE FROM THIS
Under Kevin Warsh, the Federal Reserve’s message has changed, but the constraints it faces remain the same, and the market has drawn the right conclusion. We maintain our conviction about gold’s trajectory and the level it can reach; the 16 September decision has pushed back the timing without calling the outlook into question. Some investors built up their exposure to gold equities in a very different monetary environment. For them, the current period is an opportunity to review the weight of that exposure in their portfolio: the central bank is taking a firmer tone, while in our view its room for manoeuvre in the face of inflation remains limited. The appropriate weight depends on each investor’s circumstances, but the reasons to review that allocation have multiplied since August.
DISCLAIMER
This document has been prepared by APIS Asset Management for information purposes only. It does not constitute an offer to subscribe, a solicitation to invest, or investment advice, and shall not be construed as such. The analyses, opinions and estimates it contains reflect the judgement of their authors as at the date of publication and are subject to change without notice. Past performance is not indicative of future performance. Investments in precious metals and related financial instruments involve risks, including market, volatility, liquidity and capital loss risks. This document does not take into account the particular financial situation, investment objectives or specific needs of any recipient. It is the reader’s responsibility to conduct their own analysis and, where appropriate, to consult their professional advisers before any investment decision. Reproduction or distribution of this document, in whole or in part, is prohibited without the prior authorisation of APIS Asset Management.
SOURCES
• Federal Reserve, Kevin Warsh, In Our Time, remarks at the Jackson Hole Economic Policy Symposium, 28 August 2026
• Federal Open Market Committee, policy statement and Summary of Economic Projections, 16 September 2026
• Federal Reserve, H.15 Selected Interest Rates, for the effective federal funds rate
• US Department of the Treasury, par yield curve rates, for the two-year yield
• US Senate Committee on Banking, Housing, and Urban Affairs, confirmation hearing of Kevin Warsh, 21 April 2026, for his remarks on trimmed-mean measures
• Federal Reserve Bank of Dallas, Trimmed Mean PCE inflation rate, July 2026 data, release of 26 August 2026
• Federal Reserve, H.6 Money Stock Measures, for M2
• US Department of the Treasury, statement on liquidity support buyback operations, 19 August 2026
• European Central Bank, The international role of the euro, June 2026, published 2 June 2026
• Newmont Corporation, second-quarter 2026 results, 23 July 2026
• US Congressional Budget Office, for federal interest expense
• APIS Asset Management, market update of 17 August 2026
• Bloomberg, internal extractions as at 22 September 2026: gold spot (XAU Curncy), silver spot (XAG Curncy), two-year Treasury yield (USGG2YR), effective federal funds rate (FEDL01), NYSE Arca Gold BUGS (HUI Index), Newmont (NEM US Equity) and Brent (CO1 Comdty). Headline and core PCE price indices. One-year forward EV/EBITDA and net margins for the HUI, MSCI World sector indices and S&P 500, as at 22 September 2026, forward figures based on consensus estimates