The biggest risk after a calm decade isn’t volatility. It’s forgetting that volatility is normal. In May the JSE Top 40 touched a record above 130,000 points. By the end of June it had fallen back below 110,300, and the index finished the first half of the year down 5.6%. Many investors experienced that decline as though something unusual had happened. In truth, the unusual part had been the calm that preceded it. That is the pattern worth examining. Not the volatility itself, but the surprise investors felt when it arrived. Markets don’t need to become riskier for investors to feel less prepared for them. They simply need to stay calm for long enough, and for much of the period between the 2008 financial crisis and 2022, that is about what happened. There were real disruptions inside that stretch, the 2020 pandemic crash chief among them, but ultralow rates and aggressive policy support pulled markets back to calm quickly enough that the disruptions came to feel like brief interruptions rather than evidence of the underlying regime. Risk didn’t disappear during that period. It became unfamiliar. And unfamiliar risk always feels larger when it eventually turns up. The more recent an experience, the more heavily we weight it when imagining what comes next. A decade or more of relative calm does not reduce the statistical probability of a sharp drawdown. It reliably reduces how prepared investors feel for one. It disguises itself as experience rather than error. No-one feels as if they are ignoring risk when markets are calm; they feel as if they are simply reading the evidence in front of them. Behavioural finance has a name for this: recency bias. The distinction underneath this is between statistical risk and experienced risk. Statistical risk is what history says can happen. Experienced risk is what you have personally lived through and therefore believe is normal. The two are often very different, and when they diverge for long enough, portfolios are built for the experience rather than the statistics. Diversification starts to feel unnecessary, because it hasn’t recently proved its worth. Position sizes drift higher, because recent history suggests they can. The portfolio slowly adapts to the recent past instead of the full range of futures that remain possible. Behavioural finance has a name for this: recency bias. The distinction underneath this is between statistical risk and experienced risk. Statistical risk is what history says can happen. Experienced risk is what you have personally lived through and therefore believe is normal. You could see the mechanism at work this year in individual shares, not just the index. Choppies traded above R8 at the start of 2026 on the strength of years that had rewarded shareholders. It now sits near R1.40. Mantengu climbed from 50c to R2 and fell back to about 30c, all within three years, prompting complaints of manipulation to the JSE and the Financial Sector Conduct Authority (FSCA). In both cases the recent trend had quietly become the story investors believed — right up until it wasn’t. Sasol showed the same mechanism running the other way. After years of disappointing investors the share rallied more than 50% this year as oil, coal and chemical prices strengthened due to the Iran war. Investors who had stopped expecting Sasol to surprise them were caught off guard for the same reason Choppies holders were. Recency bias doesn’t only hide risk. It hides reversals too. Whatever has been true recently starts to feel permanently true, in either direction, until the market insists otherwise. I see this less in market data than in conversations with clients. Someone who started investing during the calm years will almost always describe their own risk tolerance as higher than someone who began during a difficult market, even when their financial circumstances are nearly identical. Neither is wrong about how they feel. But neither is really describing their capacity for risk. They’re describing their memory of it. Part of the adviser’s job is separating the two, because a portfolio built on borrowed confidence from a good run isn’t built on anything durable. None of this is a forecast that markets are about to become permanently more volatile. Anyone claiming to know that is simply running recency bias in the other direction. The point is narrower, and more useful. However the next year may unfold, the job is not to guess which regime we’re in. It’s to build a portfolio that doesn’t depend on guessing correctly. In practice that means treating volatility as a recurring feature of markets rather than an occasional intrusion into an otherwise calm system. It means sizing positions so that a sharp reversal, whenever it comes, is survivable rather than catastrophic. And it means resisting the pull to quietly redefine “normal” every time markets hand you a new run of good or bad years, because that redefinition is exactly how the last decade’s calm became something investors mistook for a permanent feature of markets rather than a temporary one. The temptation after a decade like the last one is to mistake experience for probability. But experience is only one path the market happened to take, among many it could have taken instead. Not what the last decade taught us to expect, but what our memory of it has quietly persuaded us is normal. • Luthuli is investment management director at Luthuli Capital.