What the Mountain Moved

At about 8.40 on the morning of 26 August, a section of glacier gave way on the north side of the Langtang range. The debris entered the Lhende Khola and dammed it roughly 20 kilometres above the Nepal-China border; when the blockage broke, the Trishuli rose by as much as nine metres in thirty minutes. A week later, officials reported more than 1,300 dead and over 5,500 still missing.

The economics begins with a distinction that most coverage collapses. Finance Minister Swarnim Wagle has put preliminary rebuilding costs at roughly Rs 762 billion, around $5 billion. Against an IMF projection of $48.1 billion in nominal GDP for FY27, that is 8 to 10 per cent of a year's output. But this is destroyed capital stock, not forgone flow. GDP counts the latter, which is why reconstruction will eventually register as growth even though the country is poorer. Bastiat's broken window is not a curiosity here; it is a live risk of misreading the recovery.

The concentration of losses is the more interesting story. Damaged hydropower projects account for more than 12 per cent of national generating capacity, and 14 hydropower and solar projects totalling around 748 megawatts were affected. The 111MW Rasuwagadhi project was washed away entirely and Chilime was buried. Developers clustered along one steep valley because the gradient was profitable, producing precisely the correlated exposure that insurance markets handle worst. Commercial insured losses are estimated above NPR 20 billion, about $132 million, against economic losses approaching $2.56 billion: a protection gap of roughly 95 per cent.

Trade offers a warning from recent history. The Rasuwagadhi customs office was destroyed, with an estimated 100 to 200 loaded containers presumed missing. After a smaller flood on the same river in July 2025, trade through the crossing fell 95 per cent over six months, from $302 million to $15 million. Trade routes exhibit hysteresis. Contracts move, relationships re-form elsewhere, and the volumes do not simply snap back when the bridge reopens.

The binding constraint on recovery is probably not finance but absorption. Nepal executed only 12.1 per cent of its FY26 capital budget in the first half of the year, and the IMF projects central government capital expenditure at just 4.8 per cent of GDP in FY27. A $5 billion programme therefore demands a step change in procurement, land acquisition and clearances, not merely donor pledges.

Finally, the hazard is not being priced. Glacial instability is a foreseeable cost of production in high-Himalayan power generation, yet it enters no licensing test, no tariff and few balance sheets. Four flood monitoring stations were themselves destroyed. Parametric cover, contingent credit lines and hazard mapping in project approval are cheap relative to a tenth of GDP.

The rebuilt bridge will look like recovery. Whether it is depends on what it is built to survive.