Good morning. The Nvidia correction continues. Shares in the chipmaker were down 9.5 per cent at closing yesterday. Should CEO Jensen Huang trade his signature leather jacket for a cheap, all-weather denim? Email us: robert.armstrong@ft.com and aiden.reiter@ft.com. The S&P 500 fell by more than 2 per cent yesterday. The Financial Times story (and indeed every story) on the sell-off linked the decline to weak survey data from the manufacturing sector, which “added to investor concerns about an economic slowdown”. The survey-based ISM Manufacturing index for August, at 47.2, was barely improved from its dreary July level. The index’s new orders component, thought to be a leading indicator, fell almost three points, to 44.6. Remember (who would blame you for blocking it out?) that the poor July reading was the initial tremor that, when combined with a bad jobs report, became the market mini-quake of a month ago. So this seems like a good moment to ask how much we should worry about weak manufacturing ISM readings. And the answer is: not much. A few obvious points to start. First, the August number is not really news. The July and August ISM readings are not that much worse than the numbers from the past couple of years. Yes, the new orders number is a multiyear low, but it’s just one month: Next, it is not obvious that the market sell-off had all that much to do with the manufacturing report. The most notable decliner yesterday was Nvidia, off almost 10 per cent. Nvidia is not a manufacturer, and is not particularly sensitive to the cyclical trends that affect manufacturers. Broadening the point, the persistently soft signal from the ISM survey has not been picked up by the stock market: industrial stocks have done just fine in the past couple of years. In theory, we should care about a manufacturing survey — despite manufacturing being a bit more than a tenth of US GDP, and an even smaller share of employment — because manufacturing can be a leading indicator of demand in other sectors. “If there is no appetite for manufactured goods from services, retail, mining or construction, that tells you something,” says Skanda Amarnath of Employ America. That said, Amarnath also points out that US manufacturers serve global customers, so the ISMs may reflect weak global, rather than US demand. And uncertainty about the rate cycle and the presidential election may be shifting demand to the future, rather than destroying it. If the weak manufacturing ISM reading reflected low investment level by domestic customers, we would expect to see that in the GDP data on investment. But real investment growth has only fallen a bit (and investment in equipment is rising): Matthew Martin of Oxford Economics generalises the point: “soft” indicators of sentiment such as surveys and “hard” indicators of production have come apart in manufacturing. This should not come as a surprise to us at a moment when consumer sentiment (lousy) and consumption (just fine, thanks) are not tracking each other either. He thinks manufacturing is set to recover. Inventory levels are low and will have to rise at some point; lower interest rates should support investment; getting the election over with one way or the other will help confidence; and subsidies and other Biden administration supports for domestic manufacturing continue. Housing in America is punishingly expensive, and the presidential candidates have noticed. Kamala Harris has promised to increase housing supply by 3mn units. Donald Trump has pledged to bring down prices and to “protect America’s suburbs”. If the next president is lucky, falling interest rates will loosen the market. But a real fix requires new supply — and the supply side of the US housing market is mostly governed locally, not federally. Local zoning laws tend to focus on maintaining property values and a neighbourhood’s “feeling”, and thus often limit new buildings through lengthy approval processes or restrictive policies. So how can the federal government increase supply? Experts identify four policy routes: And reforming zoning laws is hard even when change is initiated locally. From Kate Nelischer at the University of Southern California: In short: there are things the federal government can do, but local zoning remains the key determinant, and the politics are tricky. Any change emanating from Washington will be painfully slow. (Reiter) Beijing in Albany.