Economic impacts of energy transitions: insights from a new economic model
In this blog, Andrew Jackson summarises a recent paper co-authored with Tim Jackson exploring the economic effects of different energy transition pathways. Using a new macroeconomic model, TranSim 2, the paper examines how key features of the transition—such as green investment, energy efficiency, and financial risks—interact in complex and sometimes disruptive ways, offering fresh insight into the challenges and trade-offs of reaching net zero.
Blog by Andrew Jackson
As the world accelerates its shift toward net zero, policymakers face critical decisions about the economic consequences of different energy transition pathways. Will a rapid push for renewables drive inflation, or will falling clean energy costs reduce energy prices? Will green investments increase employment, or will the transition away from fossil fuels generate an economic downturn or financial instability? In what way might inequality be impacted? And how might any potential impacts play out and interact with each other along different transition pathways?
Answering these questions is far from straightforward. The economic impacts of energy transitions are complex and multifaceted, and the way they are studied can significantly shape the conclusions drawn. Indeed, different papers on energy transitions, which are often based in different literatures, can end up focusing on very different aspects or features of the transition process. As such, they often provide different and even conflicting narratives around the potential impacts of energy transitions.
For example, one feature of energy transitions which has received a lot of attention relates to the impacts from green investment. One of the main narratives in this literature is that increases in green investment will generate economic growth and other co-benefits, including improvements in productivity, public health, and energy security. However, there is also a second narrative that exists in this literature which contends that green investments will ‘crowd out’ other, more productive, investments, and so lower economic growth.
Conversely, the field of net energy analysis focuses on the ‘supply side’ of the energy transition by relating technological differences between energy technologies to energy prices and economic growth. Here, the concern is that declines in the efficiency—the ‘energy return on energy invested’ (EROI)—of the energy system will generate increases in energy prices and so inflation, with potentially negative impacts for economic growth. This literature, however, is rather contested, with some more recent empirical analyses suggesting that the differences between the efficiency of fossil fuel and renewable technologies might be smaller than originally thought.
Additional points of difference can be found in the various finance related literatures. The financial transition risk literature, for example, examines the potential for certain types of transitions to ‘strand’ fossil fuel assets, which may then negatively affect the market values of these companies and even their ability to repay their debts. This may then negatively impact on broader credit conditions and financial stability. On the other hand other finance based literatures focus on the availability of credit for green investment, and the impact of changes in credit conditions on the viability of the transition as a whole.
There are, therefore, several literatures or narratives that exist around energy transitions, each of which focuses on different dynamics and potential economic impacts. And the narratives listed above are really just the tip of the iceberg—other areas of study include trying to understand the economic impacts from critical minerals shortages, the implementation of transition policies, increases in physical risks, and changing geopolitical dynamics, amongst others.
One ‘problem’ with these literatures is that they tend to operate independently of each other. In some cases this is because different the literatures are interested in different types of transitions—the transition risks literature, for example, is more focused on fast, disorderly transitions, whereas the green growth literatures often assume a slower, more orderly transition process.
Other times the focus is limited due to the paper’s scope, or because of limitations in the methodological approach. Energy transition models, for example, don’t necessarily model every potential transition narrative or feature. A good example of this is Integrated Assessment Models (IAMs), which don’t include financial institutions and so can’t model finance related features (e.g. transition risks, green financing) and their interactions with other transition features. Conversely, other models that do include financial aspects often do not account for the technological differences between green and fossil fuel energy and so are unable to simulate changes in EROI. Thus, most energy transition models do not simulate all of the potential features of an energy transition.
Consequently, the potential links and feedback effects between different transition features are under researched, as are any potential non-linear transition dynamics which might occur due to different transition features impacting the economy at different points during a transition.
In order to address this perceived research gap we developed a stock-flow consistent, input-output macroeconomic model (TranSim 2) that is able to simulate a large number of the features associated with different types of transitions to net zero. In particular, we use TranSim 2 to understand the potential links and feedback effects between different transition features and how they might generate non-linear dynamics along the transition pathway.
Simulation results
Our simulations show a number of interesting results.
First, we show that faster, more ‘disorderly’ transitions are more disruptive than slower, more orderly transitions, and that transitions in which the expectations of fossil fuel firms and investors on financial markets react more slowly to the transition are more disruptive than those in which they adjust more quickly.
In general, the faster transitions are more disruptive because they generate more inflation, loan defaults, interest rate rises, asset prices fluctuations, changes in the functional income distribution, and reductions in EROI than their slower counterparts. Faster transitions also have more of these impacts happening closer together, which magnifies the overall impact.
Conversely, the transitions in which expectations are slower to react tend be more disruptive for financial markets (i.e. for asset prices, loan defaults, and interest rates) than they are for the real economy. This is because ‘incorrect’ expectations cause fossil fuel firms to overinvest and so borrow more from banks and financial markets, and because investors on financial markets to allocate too much of their wealth to fossil fuel securities.
Interestingly, each transition can also be characterised by a number of stages. The first phase of each transition is characterised by an increase in output due to the increase in green investment, which in turn positively impacts on inflation and lowers EROI. In the second phase the higher levels of inflation combine with a slowdown in investment expenditures and a worsening of inequality to generate a downturn in consumption and economic growth. Then, in the third phase economic output partially recovers before the economy enters a prolonged period of stagnation due to the high levels of inflation and inequality. In the fourth phase there is a further downturn as financial transition risks hit the economy as the fossil fuel sector exits the model and defaults on its remaining debts. Finally, in the fifth phase the economy starts to recover and returns to growth.
Policy
There are three clear policy implications stemming from the simulation results, none of which are novel. First it is important to ensure that there are no delays in the transition process, to ensure the transition occurs as quickly as possible without generating any unnecessary transition risks. Second, the transition pathway should be clear and credible, so that firms and financial market participants invest appropriately and avoid the transition risks stemming from inaccurate expectations. Third, policymakers should incentivise the most efficient green technologies, to avoid the inflationary impacts from larger reductions in EROI.
However, it may be that a more economically disruptive transition may be unavoidable. If this is the case, the government could offset some of the negative impacts by subsidising energy prices to reduce the inflationary impacts of the transition. Alternatively, it could reduce energy prices by undertaking some of the green investment itself and operating the capital at cost (or even a loss). A third option would be to act to reduce green investment financing costs, either by derisking green investments or financing the investment expenditures itself (or through a national investment bank or the central bank). In addition, the government may also want to engage in countercyclical policy to prevent the economy from falling into recession or redistributive policies to address changes in the functional income distribution. Indeed, the changes in the functional income distribution during the transition provide a justification for green new deal type policies in which transition policies are combined with redistributive elements.
Summing up
Of course, drawing policy conclusions from economic models is a dangerous game—as George Box famously said, all models are wrong (but some are useful). It is, therefore, important to take the model results and the resultant policy conclusions with a pinch of salt. However, at the same time we shouldn’t ignore what is ‘useful’ about our energy transition model either. And what is useful about this work is it shows how different transition elements—green investment, green investment financing, changes in EROI, and financial transition risks (amongst other things) – might interact in unexpected ways along different transition pathways. Thus, models that study energy transitions should at the very least be able to include these key transition features, as a failure to do so will bias both their results and the policy advice they are able to give. Models, it seems, can be both more and less wrong and so more and less useful.
Related links
