A new question behind the dollar outlook
Federal Reserve Chair Kevin Warsh discussed emerging technology and productivity at a conference in San Francisco on 16 November 2026. He said Fed staff were examining illustrative economic scenarios, explicitly distinguishing them from forecasts. [1]
AFX analysis: for South African traders, the useful question is how changes in productive capacity might eventually affect inflation, interest rates and capital flows. This is a different horizon from a news candle. A technology speech cannot, on its own, establish the next direction of XAU/USD or USD/ZAR. This article, published on 17 November, separates the confirmed remarks from an educational framework for interpreting them.
Verified facts: what Warsh said
Warsh said recent productivity growth had been reasonably strong but had not yet shown a definitive structural breakout. The scenarios he outlined ranged from broadly shared productivity gains to growth accompanied by job displacement, or disappointing technology returns followed by weaker activity. He described the Fed’s role as preparing for potential outcomes rather than advocating for or opposing the technology. [1]
AFX analysis: that distinction should shape how the speech is used. It establishes that policymakers are considering uncertainty. It does not provide a probability for a particular outcome, a timetable for an AI boom, or a ready-made trading signal. This article does not attribute a specific market move to the speech without independently verified evidence.
Productivity means more than working harder
The Bureau of Labor Statistics explains labour productivity by comparing output growth with growth in hours worked. Its broader total factor productivity measures relate output to a combination of inputs. BLS also notes that productivity can reflect machinery, workforce skills, management, work processes and technological advances. An improvement is therefore not automatically evidence of AI’s contribution alone. [2]
Educational example: suppose a business produces 100 identical units in 100 working hours, then produces 110 comparable units in the same hours. Output per hour rises by 10%. That is a productivity improvement in this simplified example. Producing 110 units by using 110 hours would leave output per hour unchanged.
Educational framework: the example deliberately holds product quality constant. Real economic measurement is more complicated, which is why a claim about productivity should identify the measure and period being used. A software demonstration, a rising technology share price and measured economy-wide efficiency are three different kinds of evidence.
Why efficiency does not automatically mean rate cuts
AFX analysis: producing more with the same resources can ease some cost pressures. But the inflation outcome also depends on demand, wages, other input costs and how businesses set prices. More efficient production does not require every saving to be passed immediately to customers, nor does it eliminate unrelated price shocks.
Consider another hypothetical calculation. If output per hour rises 10% while hourly labour compensation rises 5%, labour cost per unit changes by 1.05 divided by 1.10, minus one: approximately negative 4.55%. This isolates one cost component. It does not predict consumer inflation or account for electricity, materials, financing and profit margins.
AFX analysis: investment needed to build new capacity can also create demand before its efficiency benefits arrive. Timing matters. A long-run supply improvement and near-term spending pressure can coexist. Treating all technology progress as an immediate reason for lower rates skips that distinction.
Verified policy context: the Fed’s 28 October statement held its target range at 3.75%–4.00%, described inflation as elevated and said more information would help determine the appropriate policy stance. The vote was 10–2, with the dissenters favouring a quarter-point increase. That remains a dated policy decision, separate from the 16 November speech. [3]
Gold: distinguish lower inflation from lower real yields
AFX analysis: one possible mistake is to assume that lower inflation must lift gold. The relevant market response is more complicated. If nominal bond yields remain unchanged while expected inflation falls, the expected inflation-adjusted return rises. That could increase the appeal of interest-bearing assets relative to gold.
An illustrative approximation makes the distinction clear. A nominal yield of 5% with expected inflation of 3% implies an approximate expected real yield of 2%. If expected inflation falls to 2% while the nominal yield stays at 5%, the approximation rises to 3%. These are invented teaching inputs, not current Treasury yields or inflation forecasts.
AFX analysis: if nominal yields instead fall sufficiently, the result can be different. Gold can also respond to currency changes, portfolio demand and uncertainty. The task is to identify which mechanism is operating, rather than attach a permanent bullish or bearish label to productivity. No price target follows mechanically from these examples.
USD/ZAR: the relative outlook matters
AFX analysis: an improvement in the expected return on US investment could attract capital and support the dollar. Alternatively, a change that reduces expected US interest rates might weaken one source of dollar support. The direction depends on the combination of growth, inflation and investment expectations, not on the technology headline alone.
For South African readers, USD/ZAR also contains a domestic story. A change in American prospects does not erase South African policy expectations or local developments. A stronger global dollar can lift USD/ZAR even when nothing new has happened at home; a local improvement can partly offset that external pressure.
Educational framework: compare the rand’s move with several other currencies before calling it a purely South African reaction. Then separate the observation from the explanation. “USD/ZAR rose” describes a price change; “it rose because of AI” is a causal claim that needs additional evidence.
How to turn the debate into a useful watchlist
Educational framework: keep three columns in a research journal: verified releases, your interpretation, and the evidence that would disprove it. For productivity, record the actual measure rather than an enthusiastic headline. For policy, distinguish a committee decision from a speech, a staff scenario or a private forecast.
AFX analysis: a stronger efficiency thesis would need repeated supporting evidence, not a single observation. Equally, disappointing technology investment would not automatically settle the monetary-policy response. Ask whether the new information materially changes your original assumptions before changing a market view.
Educational framework: match the horizon of the evidence to the horizon of the trade. A multi-year productivity argument offers little precision about a one-minute entry. If using short-term setups, assess execution costs and invalidation separately. The existence of a compelling economic narrative does not make leverage safer.
Educational risk disclaimer
This article provides general education and conditional analysis, not personalised financial advice or a recommendation to buy or sell. Forex, gold and leveraged CFDs can cause substantial and rapid losses. Stops may fill at worse prices than intended. Scenarios, forecasts and correlations can fail; no return or trading outcome is guaranteed.
