By conventional logic, low oil and gas prices suppress demand for low-carbon alternatives, and conversely high oil and gas prices spur substitution.  There is some debate that the current low oil prices will slow the transition to low carbon energies.  I argue below that we should not be worried about low fossil-fuel prices undermining the energy-transition, but rather the complete opposite.  High fossil-fuel prices will be a bigger problem.

In corporate finance the concept of the Weighted Average Cost of Capital  (“WACC”) is well known.  This is the total cost of how the company funds itself through equity and debt.  Simply put, the cost of funding a corporate entity is the percentage of equity times by the cost of equity and the percentage of debt times by the cost of that debt, thus:

  • WACC = (%E*costE) + (%D*costD)      where %E+%D=100%.

When thinking about the concept of embodied or embedded energy I have adopted and adapted the idea of the WACC to better explain the issues.  In this new model I call the Weighted Average Cost or Energy – I am trying to capture, albeit very simplistically, the idea that making wind-farms and solar panels requires a large energy input, and the cost of that energy input(“cRE”) is today dictated by the cost of fossil-fuel power (“cFF”).  This statement is true given that roughly 85% of the world’s primary energy comes from fossil fuels – so it stands to reason that this will underpin the cost base. Thus, the WACE can be expressed:

  • WACE = (85% * cFF) + (15% * cRE)

If you are constructing a renewable energy project today, the WACE for your project will be dominated by the 85%, that is by the cost of fossil-fuels.  This has important implications for the Energy Transition: if the cost of fossil fuels increase, the cost of renewable energy increases also.

From WACC to WACE

Generally, when I am discussing energy – audiences “get” the idea of embodied energy, but not necessarily the implications.  Indeed, I don’t think I know many of the implications as there may be complexities and non-linearities in the system.

However, when I hear that “renewables are deflationary” I struggle to understand.  I think this idea is argued along the lines that because the price of solar panels or wind farms has dropped massively in recent years (which indeed they have due to technological improvements and economies of scale), so the cost of energy decreases and so they make everything cheaper.

If so, this is a reasonable concept – given that this is exactly what happened over the last 100 years as hydrocarbons started to power the world economy.  This can be illustrated by the fact that household disposable income rose dramatically – the result of the “essentials” becoming significantly cheaper. 

For the last 75 years at least, the price of goods and services has been highly correlated to the price of oil.  This is the effect of embedded energy and it is as relevant for food as it is for industry.

A screenshot of a cell phone

Description automatically generated

I don’t see this relationship changing anytime soon: let me explain.   Whilst we see dramatic headlines and charts of the growth of renewables, the reality is that thirty years ago (1990) the world got roughly 87% of its primary energy from fossil fuels, but 2020 that number was about 84.7% (BP Statistical Review).  These figures vary depending on the source, but the quantum is the same – at a global level, we are simply using more of everything, and the proportion of fossil fuels in the energy mix has declined only very marginally despite the huge efforts to transition.  In the BP data there is a noticeable acceleration in the most recent years, although this does not always show in other data.  The rate of increase in penetration is an important variable in the subsequent discussion

So, let us examine the future in terms of a thought experiment using the Weighted Average Cost of Energy

Ever Cheaper

A picture containing indoor, building, table, many

Description automatically generated
Heavy Lifting

If renewable energy is systemically cheaper that fossil fuels, we should see a positive feedback loop setting up.  Remember that whilst renewable energy is renewable, the machines that harvest it are not.  These machines will need to be replaced over and again at some interval.  There is a monetary and energetic cost in making these machines.  Today, the world is powered roughly 85% by fossil fuels.  It is not unreasonable therefore to assume that 85% of the cost of building a wind farm or solar panel is underpinned by the price of fossil fuels.  Of the 15% that is not fossil-fuels (“FF”), it is dominantly nuclear and hydro, but lets park that for a moment and lump all “Low carbon” together as Renewable Energy (“RE”), this is quite aggressive as only 4% is actually Solar and Wind: Hydro and Nuclear are not increasing their market-share.

In reality, due to the heavy industrial processes required the proportion of fossil fuel energy is likely higher than the global average.  But park that also.

So, our starting point is: WACE = ((85% * cFF) + (15% * cRE))

In this “Ever Cheaper” scenario we are told Renewable Energy is as cheap as fossil fuels and rapidly getting cheaper still.  Thus, we should expect two things: (1) the cost of renewable energy  (“cRE”) to quickly break below the cost of FFs, and (2) the relative percentages should increase as the economics will drive the transition in addition to policy.  We could add a third assumption, that the rate of change of will increase (accelerated transition) as the positive loop feeds back.   I have not modelled the latter, but simply taken a linear increase in the penetration, possibly underestimating the outcome.   We will also ignore Jevons Paradox, which would suggest that increasingly cheap energy would increase demand.   With this in hand we can think about how the average cost of energy changes over time as this is an iterative process.

This is quite easy to model with a few input assumptions – and the result is perhaps obvious: notably, the weighted average cost of energy decreases.  The conclusion is that society will have ever cheaper energy (which happily will, in this version of the future, be greener and cleaner). 

In the above example, the input assumptions are (a) Renewables increase their market share by 9 percentage units at each step achieving 98% penetration by step 10, (b) fossil fuels decrease in cost also by 5% at each step (due to decreased demand) and (c) the cost of renewables decrease by 30% at each step.  Clearly this is an extreme example, but by pushing the logic, one starts to see the absurd outcome – a deliberate form of reductio ad absurdum.  After 10 iterations, the WACE is only one quarter of the starting WACE and heading to zero.  Free Energy.

This is not presented here as a reasonable or probable scenario, but simply included as to illustrate the paradox, in full acknowledgement that it is “extreme”.  The 30% compounding reduction in cost of Renewable Energy at every step is clearly unrealistic.  Given the huge cost decreases to-date, and the known physical limitations on efficiency, we probably can’t expect such huge decreases in future.

The other side of this coin, as noted in the first part of this essay, is that cheaper energy will increase purchasing power and drive consumerism – which may be a bit of an own-goal in the bigger scheme of things.  But again, lets park that thought since it isn’t going to happen.

Fundamentally this thought experiment leads to the conclusion that energy will become, after a certain number of iterations, essentially free.  If that is absurd, what is the alternative?

Ever More Expensive

I personally don’t think that ever cheaper energy is possible – this conclusion kind of rubs up against thermodynamics in an uncomfortable way.

A screenshot of a cell phone

Description automatically generated

The opposite scenario is one which I believe is much more likely, more immediate and more problematic.  In this case we can model what happens if fossil fuel energy gets progressively more expensive.  This may seem a bit odd given that oil went negative a few weeks ago.  But ignoring the highly unusual market conditions of the Covid pandemic, the reality is that humans have been very good at exploring for oil and gas and that the best deposits have been exploited. 

Peak conventional oil production was, by some reckonings sometime between 2005-2010.

Oil-sands and shale-oil are both forms of unconventional recovery that are intrinsically more expensive, Deep-water is also sometimes included in this category – and it is these that have plugged the demand gap for the last 15 years.  If shale oil was indeed the “marginal barrel” in recent years, then that points to a cost of north of $65/bbl (WTI) – cutting through all the smoke and mirrors of the creative accounting.  For an independent operator to make money after all costs (corporate overheads, G&A, financing costs – as well as capex and opex), that is the ball-park price needed.  Even adjusting for inflation this is higher than the historical cost of oil.  Oil is getting more expensive as we use up a finite resource.  As discussed in previous posts – high oil prices are bad for the global economy.

The “worst case” scenario could be thought of as slowly increasing costs of fossil-fuels, and a slow transition.  This will increase the WACE and starts to look quite concerning.

For this I have set the penetration of RE to increase by 1% point per step (recall that RE has gained about 2% in the last 30 years), the cost of FF increases 5% per step and the cost of RE decreases by 10% per step.

A close up of a map

Description automatically generated

The result is a worrying increase in the WACE over time – again, higher energy costs make everything more expensive, so this is not good news.

It could be argued that the transition to renewable energy has happened much more quickly in places like the UK.  However, the fact that somewhere like the UK has transitioned its electricity supply much more rapidly is pretty much irrelevant.  Firstly, the transition has happened on electricity, not on all primary energy.  Electricity makes up roughly 20% of primary energy.  Secondly, the UK does not make many (any?) industrial scale wind or solar.  Thus, the cost-base that is being assumed as an input variable must be established from where these are mined / processed / transported / manufactured / re-transported etc. not some local hot-spot of renewable electricity. Therefore, it is appropriate to use the global WACE not a local WACE.

Why will Renewable Energy get more expensive?

As noted previously, Renewable Energy has to be harvested with large machines.  These have embodied energy, the cost of that energy input is the WACE at the time of fabrication.  If the dominant part of that equation (cFF) increases due to increasing scarcity, so the WACE for new harvesting machines will increase. 

A close up of a piece of paper

Description automatically generated

We can assume that there are further cost savings to be had due to improvements in technology and further economies of scale in the industrialization of RE.  Thus at each step the cost base increases (due to cFF inflation) and is reduced by some “improvement” factor for cRE.  In this example the inflation of cFF is 5% per iteration, and the resulting cost input for RE is then reduced by 20% at each step.  In addition,  RE penetration increases by 2% per step.   As you can see, this the cost of RE increases incrementally – this is driven by the modeled increase in the cost of fossil fuels over-riding the significant declines in the cost of the Renewables, simply due to the mismatch in relative proportions in the overall energy mix.

Now one could argue that the spectacular declines in the cost of wind and solar will continue, so the increasing costs of the fossil-fuels parts of the manufacturing/transporting/installing processes will be offset (or more than offset) by further declines due to technology and/or economies of scale.  Whilst this may be possible, it is less likely now that we are already far along the innovation curve.   There are very clear limits to this, and most of the improvements in both technology and scale have been achieved. 

There will be incremental improvements, but step-changes are unlikely.  Unfortunately, and in contrast to some of the more strident evangelist’s positions, this is not a digital technology and will not follow Moore’s law.  Moreover, the declines in unit costs of RE machines does not translate into declining cost of energy on a like-for-like basis.  This is a whole separate subject (Levelized cost of Electricity) – and is way beyond the scope of this essay.  But on a like-for-like basis RE+Storage (where the “storage” part is technologically out of reach source), the cost or renewable energy at a societal level provides more expensive electricity.  So rather than modelling the decrease in unit costs, maybe one should model the increase in electricity prices into the WACE?

Supply Shock Scenario

I noted above a “worst case” scenario – and deliberately used quotation marks.  The quite plausible scenario is that we will have an oil-price supply shock – following on from, and indeed partly caused by, the current low oil price situation.  Let’s see what happens if the cost of fossil-fuels jumps.  In this case I assume a fairly modest 50% increase in the cost of fossil fuels, and no subsequent slump – but no further increases either.  A 2% step-wise increase in penetration rate and a 5% step-wise decrease in costs of renewables.  Due to the weighting of the 85% market share, this 50% increase dominates the picture

A screenshot of a cell phone

Description automatically generated

As an exercise I looked at what sort of cost savings would be needed on renewables to offset this increase in fossil fuels – and with a 2% increase in penetration, we can see that a very extreme case of 95% decrease per step (ie with RE components being zero cost after a couple of iterations) just about manages to offset the spike in fossil fuel prices.

A close up of a map

Description automatically generated

But in reality it is worse than this. If I had modeled the energy mix more accurately, with 85% fossil fuels, 11% Hydro and Nuclear and only applied the cost reductions to the 4% which is currently Renewables, the outcome would have been even more extreme of course. The equation should be:

WACE = (85% * cFF) + (11% * cHN) + (4% * cRE)

Ever More Expensive – revisited

But what about the externalities that are not priced into the price of fossil fuels? 

CARBON TAXES – UNINTENDED CONSEQUENCES

Reasonable question – but by using the WACE we should understand that if we ever find a carbon price that is appropriate – the cost factor (cFF) in the equation will increase and this will push up the average cost – thus the next generation of renewables will be more expensive.  Taking a relatively middle of the road case: Penetration increasing 2%, Cost of FF increasing 5% and Cost of RE decreasing 10% each step, we could add a “carbon tax” and suggest that the cost of Fossil Fuels will increase by 7% instead of 5%.  Unsurprisingly, the WACE increases.  And to labour the point: an increasing WACE makes everything more expensive, everyone poorer and humankind less resilient.

DIVESTMENTS – UNINTENDED CONSEQUENCES

The WACC was noted in the introduction.  Large, well-loved corporations can borrow very cheaply as they are considered low risk. Equally they can raise equity at high valuations – which ensures that the dilution of existing shareholders is minimal – thus the equity is cheap. 

Conversely, a struggling company will have to give away a big chunk of equity to raise new funds (making the equity high cost) and if any debt is available it will also be high priced.  This notion is also referred to as the “Cost of Capital”. 

The global movement to “divest” from fossil fuel companies is pushing up the cost of capital for these companies.  Whilst the virtue signalling of divesting may sound right-on-message to stakeholders, this may be very bad for the global economy.  Exactly as for Carbon Taxes, increasing the cost of fossil fuels will push up the cost of everything, and that includes renewable energy.

A screenshot of a cell phone

Description automatically generated

WACE for Fossil Fuels.

Having been modelling the cost of Renewable Energy as a function of the N-1 WACE, surely this should be done for the cost of fossil-fuel energy, which also require vast industrial machines to extract the resources?  Well yes and no.  Yes, because no good or service can escape the concept of embodied energy – but possibly “No” because the cost of fossil fuels may well react to scarcity more quickly than an iterative process of industrial exploitation cost.  This is the “Supply Shock” scenario described above.  That being said, as one looks to develop more fossil fuels after a prolonged supply shock, then yes, the WACE will come into play.  But again, it is not a favourable scenario because everything conspires to increase costs over time.

EROEI

Whilst this may all seem very pessimistic, the physical sense of this can be traced back to the concept of Energy Return on Energy Invested (“EROEI”) – unconventional oil has a lower EROEI than easy oil, Renewable Energy has lower EROEI than oil.  It is hard to see how adding lower EROEI energy into the mix will reduce the cost of energy – quite the opposite.  I think the WACE is a useful tool for quantifying this. Throughout this exercise I have used the idea that Renewable Energy can be on parity or cheaper than fossil fuel energy. Despite all the limitations and flaws of EROEI, physics seems to point to low EROEI sources being more expensive on a unit energy point of view, so we should consider the case where renewanble energy is actually more expensive. In this case one should not attempt to transition as it will destroy the economy before climate change gets a look-in.

Conclusion

For those brave enough to have got this far, it may seem that I am arguing against renewable energy.  Not so fast! 

In my slightly unorthodox worldview, we are facing a threat that is more immediate than climate change – that of rising energy costs due to creeping scarcity – and in particular oil.  This is not simply Peak Oil theory revisited – there is plenty of oil.  Rather it is a concern that we have passed Peak Cheap Oil.  Worse still is the nexus of a looming supply shortfall, asset managers seeking to increase the cost of capital for fossil fuel companies and the spectre of carbon taxes.  A spike in the price of oil will apply very material limits to the global economy; limits that will make Covid19 look like a picnic. 

In contrast to the view that high oil prices are good at spurring the transition, high oil prices will make the transition to renewable energies affordable, and ultimately accelerate the collapse into expensive dirty energy.

If we accept the idea that we have only expensive and more-expensive oil left, it is entirely logical to wish for increased penetration of renewables.  This will reduce the immediate demand for fossil fuels (albeit rather more marginally than most people would expect or want) and allow the global economy to survive for a while longer on a mix of all energies. 

Dogmatically working to kill off fossil fuels (and by the same token nuclear) will lead to ruin.  We need a balanced debate and balanced energy supply.