Stranded asset and carbon pricing risk in the steel industry
Written by Matt Gray and Badr Ben M'barek
Steel is an indispensable part of society and an essential material for the energy transition. Therefore, transition risk in the steel sector requires more nuance than other parts of the energy value chain, such as upstream oil and gas, due to the expectation that steel demand will increase for the foreseeable future.
This post presents a framework and methodology for analysing stranded asset risk in the steel industry and applies this framework to our recently launched global steel cost tracker (GSCT). The objective is twofold (1) highlight the implications of inaction for conventional blast furnace and basic oxygen furnace (BF-BOF) steel production in a scenario consistent with net-zero emissions by 2050, and (2) outline why high-cost producers will use border taxes and trade agreements to engender a level playing field on carbon pricing.
What is the Global Steel Cost Tracker (GSCT)?
TransitionZero is a climate analytics not-for-profit established to clarify complexity with data transparency. We do this by developing open data and open source projects to support economic and financial decision making in electricity and industry sectors.
GSCT is an open data project that estimates the plant production costs of BF, BOF and electric arc furnace (EAF) steel production routes representing around 90% of global production. The cost estimates are broken down by raw materials, energy, labour, and other costs. GSCT is based on a methodology report by TransitionZero and Global Efficiency Intelligence and a database of steel plants developed by the Global Energy Monitor (GEM).
The truth is, only the plant operators know their exact production costs. Only by plant operators opening up, can there be a candid and proactive conversation about transition risk and opportunity in the steel sector.
Model principles
There are several definitions of stranded assets in the energy sector. The International Energy Agency (IEA) defines stranded assets as: “those investments which have already been made but which, at some time before the end of their economic life (as assumed at the investment decision point), are no longer able to earn an economic return as a result of changes in the market and regulatory environment brought about by climate policy”.
For this analysis, stranded assets are defined as the amount of conventional BF-BOF production (i.e. excluding innovative BF-BOF routes with carbon capture and storage - CCS) outside a carbon budget consistent with net-zero by 2050.
Our modelling approach is based on the following principles.
Focus on conventional BF-BOF production. In the IEA’s net-zero emissions (NZE) scenario, emissions from steel will have to decline by two thirds by 2040. We focus exclusively on conventional BF-BOF production because this process route is responsible for most direct carbon emissions from steel. The steel industry emitted around 3.6 billion tons of CO2 emissions in 2020, with BF-BOF production responsible for most of this pollution.[1]
Highlight implications of inaction. The purpose of stranded assets analysis is to highlight the implications of inaction. The intention is to emphasise how existing BF-BOF production will need to be transformed to be net zero-aligned. This transformation requires retrofits to phase-in new processes, such as hydrogen, bioenergy, direct electrification - and CCS.
Economically rational outcome. Our model optimises for an economically rational net zero-aligned world. Substantial amounts of steel are traded globally, meaning steel is a trade-exposed industry. The cost of steel production is an important factor in keeping a steel-producing country or a particular steel plant competitive in such an international market. As such, we assume the highest cost producers are forced to rationalise production first.
Existing capacity only. We exclude investments in new conventional BF-BOF capacity, as GSCT only covers existing capacity. However, there are significant investments in BF-BOF capacity being made, which are arguably inconsistent with a net-zero by 2050 outcome. GEM, for example, found $70 billion of new BF-BOF capacity could be at risk of stranding.
No carbon price assumption. This analysis does not assume a carbon price. There is currently no international carbon price covering steel emissions. Those regional carbon prices, such as the EU's emissions trading system (EU ETS), have generous free allocations to prevent carbon leakage. As discussed below, we believe an international carbon price will emerge from border taxes and bilateral agreements.
Model setup
The model setup is based on three steps.
Estimate a carbon budget for conventional BF-BOF production based on publicly available data. The Institute for Sustainable Development and International Relations (IDDRI) net-zero steel project models the changes needed to align the steel industry with a net-zero outcome by 2050. This project provides production data from existing and new iron and steel facilities from 2030 to 2050 by technology. We use the projected BF-BOF production, which maintains the energy and emission intensities of existing facilities until they are scheduled to be retired or retrofitted.
Estimate conventional BF-BOF production based on our technology. We do not forecast production into the future. Conventional BF-BOF production is based on a 2015-2021 average from TransitionZero’s technology that uses satellite imagery and data science to estimate the utilisation rate of BF-BOF facilities.
Estimate conventional BF-BOF production costs based on GSCT. As with production, we do not forecast costs into the future. Conventional BF-BOF production costs are based on a 2015-2021 average from GSCT.
132Mt per year of conventional BF-BOF production at risk by 2030
The data visualisation below is a variation on a traditional cost curve. Instead of just showing how much steel can be produced at a certain price from existing BF-BOF capacity, it estimates and overlays a carbon budget for conventional BF-BOF production consistent with net-zero emissions by 2050. In an economically rational Paris-aligned world - where conventional unabated BF-BOF producers are forced to rationalise output - 10% or 132Mt per year of conventional BF-BOF production from Japan, Germany, China, Italy and the US is outside a net-zero carbon budget by 2030. Based on a 2015-2021 average from GSCT, the production cost of 132Mt per year between now and 2030 is $1.1 trillion per year over the same period. These production costs could prove conservative, given the spike in commodity and electricity prices from Russia’s invasion of Ukraine. By 2040 this percentage climbs to nearly 40%, with Japan, Germany, Italy and the US being the most impacted. The amount of conventional BF-BOF production outside the budget by 2050 increases to 514Mt per year, with the remaining production still in operation, will likely be equipped with CCS.
Carbon pricing through the backdoor
The purpose of this post was to illustrate the implications of inaction as the world charts a path to net-zero. Despite the steel industry’s importance to the global economy and the energy transition, no carbon-emitting sector should expect a free pass to operate as usual between now and 2030. As with most heavy industry subsectors, steel is wickedly complex. While crude steel is a globally traded commodity, there are numerous nuances associated with its production due to its strategic importance. For this reason, it seems unlikely high-cost producers will reduce output without a level playing field on carbon pricing.
The findings of this analysis assume no carbon pricing and perhaps explain the recent trend to introduce carbon border taxes, such as the EU’s carbon border adjustment mechanism (CBAM). CBAM requires non-EU producers to pay the same carbon price as EU domestic producers regulated under the EU ETS.[2] If carbon intensity data is unavailable, importers will be assigned default values on carbon emissions to determine the number of certificates they need. Until free allocations end, the CBAM will only apply to the proportion of emissions that do not receive free allowances under the EU ETS.
Recent trade deals between the EU and US and Japan and US also imply trade is going to be used to support and enforce environmental goals on the international stage. Given climate change is an inevitable and urgent challenge, we see border taxes and trade agreements as an indirect way of instigating a global price on carbon. If implemented correctly, these initiatives could be an effective way to incentivise process retrofits and efficiency improvements. We hope open data, such as GSCT, can be used to start a conversation about the risks and opportunities associated with these decisions.
Notes
1. Based on our analysis, global BF-BOF steel production emitted around 3.1 Gt CO2 and global EAF steel production emitted around 0.5 Gt CO2 in 2020. EAFs in China and India extensively use pig iron or coal-based direct reduced iron (DRI) as feedstock instead of steel scrap, creating higher than usual carbon intensities for EAFs and causing an increase in global EAF CO2 emissions.
2. If the proposal is adopted without change, EU importers will have to report emissions embedded in iron, steel, cement, fertiliser, aluminium, and electricity generation from January 2023. EU importers will start paying a financial adjustment from January 2026.
