France fractures, Audi pivots, and Serbia votes: pressure on three fronts

France's debt problem is best understood as a confidence spiral. A government borrows by selling bonds, promises of future repayment, to investors. If investors doubt the government can manage its finances, they demand a higher interest rate to compensate for the risk. France's deficit, the gap between what the government spends and what it collects in tax, has been running above the eurozone's agreed limit of 3% of GDP for several years. That alone would be manageable. The problem is that a contracting economy shrinks tax revenues at the same time as protests raise the political cost of cutting spending, so the deficit gets harder to close, not easier.

The precedent that comes to mind is Greece after 2010, when a large deficit combined with weak growth and political paralysis pushed borrowing costs to levels the government could not sustain. France is not Greece: its economy is far larger, its debt is held more broadly, and the European Central Bank has tools it did not have in 2010. But the mechanism is the same. What I find genuinely uncertain is whether the French government can pass a budget that satisfies bond markets without triggering protests large enough to bring it down. If it can, borrowing costs stabilise. If it cannot, the ECB faces a choice between letting spreads widen or intervening again, and that choice has costs either way.

A tariff is a tax charged on goods when they cross a border, paid by the importer, which is usually a domestic firm, not the foreign exporter. When the US imposes a tariff on European cars, American dealers pay more for each Audi they import, and that cost is passed on to buyers as a higher price. At some point, the tariff makes it cheaper for Audi to build the car inside the US and avoid the tax entirely. That is called tariff-jumping investment, and it is exactly what this factory announcement represents. The investment does not happen because the US is the best place to build cars on pure cost grounds; it happens because the tariff changes the calculation.

The interesting question is who actually wins. Trump can point to a $4 billion factory and future jobs in Georgia. But Audi's shareholders are committing capital to a market they would not have chosen freely, which raises their costs and may reduce the quality or variety of cars they can offer. German workers and suppliers lose work that would otherwise have stayed in Germany. And the factory will not produce a single car until the early 2030s, so the near-term economic effect is close to zero. One month of announcements is not proof of a manufacturing renaissance; it is proof that tariffs change where firms invest, which is a different and more complicated thing.

Serbia's election is a good example of what political economists call a rentier state dynamic in a softer form: a government that controls enough of the economy, through state-owned firms, media and patronage networks, to make it very hard for opposition movements to convert popular anger into votes. Vucic's party has dominated Serbian politics since 2012 partly because it controls the jobs and contracts that many voters depend on. The student movement that has been running for two years is unusual because it is organised outside those networks, which is why it has lasted longer than most previous protests.

The foreign policy stakes are real. Serbia is an EU candidate country, meaning it is formally in the process of joining the bloc, but it has refused to join EU sanctions on Russia and has maintained close military and energy ties with Moscow. That position has suited Vucic because it gives him leverage over both sides. An opposition government would face pressure from Brussels to align more closely with EU foreign policy, including on Russia. The precedent worth thinking about is Ukraine in 2014, where a change of government produced a sharp foreign policy pivot and then a Russian response. Serbia's geography and size make a direct parallel unlikely, but the underlying logic, that great powers treat smaller neighbours' elections as contests for influence, applies here too.