An owner buying conventional tonnage today is financing a twenty to twenty-five year asset on debt that runs five to seven years. Two electricity transitions that have largely finished suggest value starts moving long before that debt matures, and that the credit assessment arrives later still.
A report by Marie Fricaudet, Prasanna Colluru, Nishatabbas Rehmatulla and Tristan Smith at the UCL Energy Institute examines four cases to establish the timing. Two are the German and United Kingdom electricity transitions. Two are shipping collapses that devalued assets quickly and passed the losses on, the German KG fund model and the Hanjin bankruptcy. The electricity cases carry the evidence on when stranding starts. The shipping cases show how the losses travel once it does.
German coal utilisation rates started falling in 2008, suppressed as wind and solar displaced dispatchable generation through the merit-order effect. Impairments on utility balance sheets ran from 2011 to 2016 and reached 5 percent of total asset value per year in 2016. Early closures between 2011 and 2023 produced 223 GW-years of stranded capacity, the equivalent of 22 GW retiring a decade before the end of its expected life.
Moody’s acknowledged the effect of renewables on utility credit ratings in 2012 and recognised stranded-asset risk in Germany specifically in 2014, roughly six years after utilisation began to fall. The market capitalisation of the four largest utilities never recovered from the 2008 crisis, underperforming the wider economy from 2009 onwards. Equity investors held their positions until the stranding had already materialised.
The report’s central timing finding is that stranding began around the inflection of the technology’s S-curve, when renewables held roughly 10 percent of generation and were growing quickly. Scalable zero-emission fuels and energy-saving technology in shipping remain well below that share.
Transition risk that materialises in the middle or later life of a ship falls outside the horizon of the debt that funded it, so an individual bank can refinance or exit before stranding arrives. Where that approach is general, it produces refinancing risk at the level of the system: vessels needing refinancing at the same moment that lending standards tighten create the fire-sale conditions in which recovery values fall below outstanding balances.
The two shipping cases show what that looks like in practice. The KG model raised roughly EUR 30 billion from about 440,000 retail investors and financed more than 20 percent of the international container fleet, through single-purpose vehicles each holding one ship, with debt at 60 to 70 percent of the vessel price and soft costs high enough to raise a vessel’s operating cost by up to 40 to 50 percent. The downturn that followed 2008 was general across container shipping. What separated the KG funds was the structure through which it was transmitted: single-asset vehicles with hundreds of dispersed retail investors, non-cancellable management contracts, high debt and high soft costs left almost no room to restructure. More than 150 single-ship funds filed for bankruptcy in 2012 and around 200 more in 2013.
The losses reached the lenders. Commerzbank’s ship finance division lost close to EUR 850 million in 2009. HSH Nordbank doubled its reserves to EUR 1 billion against a book of about EUR 35 billion in ship loans. Bank lending fell from 84 percent of shipping finance in 2008 to 63 percent in 2014.
Hanjin ran the same sequence at higher speed. Claims against the company totalled USD 10.5 billion against USD 220 million recovered, a payout of about two cents per dollar claimed, and owners who had chartered vessels to the line found their pre-filing charter claims ranked lowest in the Korean rehabilitation process. The creditor response changed the size of the final loss. German lenders deferred enforcement to avoid crystallising losses through fire sales, an approach that was defensible case by case and, applied across hundreds of funds, postponed and deepened the collapse. Korean lenders reclassified the Hanjin loans early and provisioned against nearly the full amount, containing the cost to around 11 basis points of risk-weighted assets.
Ownership is far more fragmented than in power generation. The United Kingdom had 58 utilities and Germany 388 with capacity above 10 MW. Shipping has close to 30,000 owners, financed by Korean, Japanese, Chinese, Singaporean and European institutions operating under different disclosure requirements and supervisory expectations. The coordinated repricing that eventually arrived in European electricity lending is harder to produce across a population of that size.
Two features work the other way, and the report is careful about how much weight either carries. Vessels can be retrofitted to accommodate alternative fuels, which power plants generally could not, though retrofit costs are substantial enough that the option reduces stranded-asset risk without removing it. Coal generation carried a structurally similar expectation, that carbon capture and biomass co-firing would extend the economic life of existing plant, and neither technology arrived at the anticipated scale or pace. Vessel lives are also shorter, twenty to twenty-five years against thirty-five to forty for power plant, so a transition of equal speed would strand less value. Set against that is the compressed timeline in the IMO strategy, which targets a 20 to 30 percent emissions reduction by 2030 and 70 to 80 percent by 2040.
The commitments that did the damage were all made while the information was ambiguous. Germany’s four largest utilities planned some 20.3 GW of new coal capacity between 2005 and 2008, while renewable capacity was already accelerating. Of that, 14.3 GW was eventually cancelled, and the 6 GW that was built became the stranded exposure. Ordering across the container market continued into an oversupplied decade. The cost of waiting for the picture to resolve fell years later, on equity holders first and on lenders after them.
Two of these cases are electricity and two are post-2008 shipping collapses with no connection to decarbonisation. Between them they establish when value started moving in transitions that have finished, and how losses travelled through structures that failed. When shipping’s own inflection arrives, and which vessels it reaches first, are different questions, and the report does not forecast either.