The Share Market Had Its Worst Day Since 2020. Gold Went Up. What Changed?

Jimmy Wong

Start with where gold was eight days ago. On Friday 30 October it fell below US$3,900 to about US$3,875, a one-year low and roughly 30% under January's record. By Thursday this week it had climbed to US$4,146, a two-week high. Friday's jobs report took about 1% back, and gold finished near US$4,050.

Over the week that is a gain of just over 4%, its best week since late August. Silver did much the same, recovering from a one-year low near US$49 to about US$52. In Australian dollars, gold came back from roughly A$5,510 to around A$5,800.

So why did gold jump mid-week? The US held its midterm elections on Tuesday, and the Democrats swept both chambers. Markets read that as two years of gridlock, which limits how freely Washington can spend and change policy. The US dollar fell to a one-month low and bond yields fell with it.

That is what matters for gold. Gold pays no interest, so it competes with bonds and cash. When bonds pay less and the dollar weakens, gold looks better by comparison, the way money leaves a savings account that has just cut its rate. On Wednesday gold rose about 3%, its biggest single day since 19 August.

Then shares fell hard, and gold held. Why is that different? On Thursday the AI trade cracked. The S&P 500 fell 4.4% and the Nasdaq 100 had its worst day since 2020. Gold still finished the day higher, though it came off its US$4,146 high as the selling deepened.

A fortnight ago we described the opposite. Shares fell, and gold was sold with them, because investors facing margin calls sell what they can sell fast rather than what they want to sell. So why not this time? Look at the bond market. In October, yields were rising while shares fell, so gold had nothing pulling it up while the cash-raising pushed it down. This week yields were falling, so gold could absorb some selling and still finish ahead. A falling share market is not automatically good or bad for gold. What the bond market is doing at the same time decides which way it lands.

Who is actually buying? The World Gold Council's quarterly report landed on Tuesday, and it is the clearest answer we have had in months. Central banks bought 321 tonnes in the September quarter, up 42% on a year earlier and the most since late 2024, with China adding 63 tonnes and Poland another 40. Gold funds reversed their earlier outflows and took in 211 tonnes, lifting holdings to a record 4,256 tonnes.

Everyday buyers went the other way. Jewellery demand fell 17% to its weakest September quarter on record, and bars and coins fell 19% to a four-year low. Picture a house auction: if the families who wanted to live there step back, the price only holds if a bigger buyer steps in. That is roughly what has been happening with gold.

Then Friday took some of it back. The US added 165,000 jobs in October against about 65,000 expected, nearly triple the forecast, and revisions added another 84,000 to August and September. Unemployment slipped to 4.1% and wages rose 0.4% for the month. The jobs report is the Fed's pulse check. A weak pulse means the patient may need help, so rates come down. A strong pulse means the Fed can hold or push higher. Traders now put the chance of a December rate rise at about 77%. Gold fell about 1%, and Goldman Sachs cut its end-2026 forecast to US$4,150 from US$4,650.

So is the one-year low the bottom? One week does not settle that, and we would be guessing if we said otherwise. What we can say is what has happened and what has not. The 30 October low has held so far, the week closed more than 4% higher, and the buyers who hold gold for years rather than weeks bought more of it last quarter, not less. What is not settled is the Fed, which meets on 8–9 December with a rate rise the market's base case, and US$4,000, a line gold has crossed in both directions over the past fortnight.

Our own reason for holding gold has not moved either way. We hold it because governments keep borrowing and money keeps losing its buying power, and neither of those is decided by five days of price. On the price itself, the levels below are US$4,000 and then the 30 October low near US$3,875. Above, the two-week high at US$4,146.

One last thing: why did the miners move so much harder? On Thursday the All Ords Gold Index rose almost 12%, and the broader materials index had its biggest day since March 2020. That is far more than gold itself moved.

Part of it is simple: a miner's costs are mostly fixed, so each extra dollar of gold price drops almost straight into profit, and that works in both directions. The second reason is one of our own observations. ASX gold shares track silver more closely than they track gold, because investors buy gold first as the cautious choice, while silver and mining shares need the next step, where investors are willing to take more risk. Silver rose 4.6% on Wednesday alongside gold, so that step happened and the miners followed. Had silver lagged, we would have expected the rally to fade. The gold-to-silver ratio still sits near 78, and a ratio that high says investors remain cautious.