andy rothman

Sharing a new piece in Foreign Policy with the fantastic Ma Xiao from PKU, unpacking some of the local govt dynamics behind China's industrial policy, and why they make Beijing's efforts to rein in overcapacity so difficult! (By the way Xiao has done, in my mind, some of the most insightful research on this topic… Add his work to your reading list if you have not already done so!) The general idea: we tend to talk about “China’s industrial policy” as if it is coherent. Many already understand that local governments have their own incentives and that implementation differs. But how do they actually bargain, compete, improvise(!), and sometimes push much further than Beijing intended? We start with two puzzles: Beijing planned for 12,000 km of high-speed rail by 2020; China ended up with roughly 40,000 km. Beijing's auto ambition was to have a small number of large SOEs and JVs. Instead, many of the real stars like Geely, BYD, Nio, etc. are private firms backed by local governments, often finding ways around regulatory restrictions with local help. Our key argument: local governments are (opportunistic) policy entrepreneurs. They have few formal ways to push their interests at the top, so they cultivate allies, bargain through ministries, back private firms, launch things first and seek approval later… and compete intensely for one scarce resource: top leadership attention! That “attention economy” is key. Hence the focus on showcase projects. And hence, when it comes to stuff that looks good on camera, e.g., EVs, drones, robotics, etc., they pile in. The fiscal angle has of course been discussed too, but is also key: Local governments still have every incentive to prop up firms, protect employment, avoid bankruptcies, etc., especially in places with few alternatives. The Shenzhens of China can afford to let a zombie EV firm fail; an already fiscally constrained city heavily dependent on one or two industries has very different options. Then there is the cadre system. Officials get credit for launching a mega project, then often rotate elsewhere before the costs / waste / downside risks show up. The average tenure of provincial and city leaders is only about three and a half years, so achievements have to show up fast. We then use this framework to explain why rebalancing toward consumption is so hard: consumption is slow-moving, difficult to measure as a local KPI, and hard to credit to a particular locality… and why reining in local debt is so difficult, since debt-led growth shows up before the consequences do. Full article below: China’s Overcapacity Has a Mind of Its Own Beijing can’t rein in its own industrial machine. foreignpolicy.com/2026/10/26…