This group brings together the best thinkers on energy and climate. Join us for smart, insightful posts and conversations about where the energy industry is and where it is going.


Are Renewables Set to Displace Natural Gas?

  • Bloomberg’s renewable energy affiliate forecasts that wind and solar power will make major inroads into the market share of natural gas within a decade. 
  • This might be a useful scenario to consider, but it is still likelier that coal, not gas, faces the biggest risk from the growth of renewables.

A recent story on Bloomberg News, “What If Big Oil’s Bet on Gas Is Wrong?”, challenges the conventional wisdom that demand for natural gas will grow as it displaces coal and facilitates the growth of renewable energy sources like wind and solar power. Instead, the forecast highlighted in the article envisions gas’s global share of electricity dropping from 23% to 16% by 2040 as renewables shoot past it. So much for gas as the “bridge to the future” if that proves accurate.

Several points in the story leave room for doubt. For starters, this projection from Bloomberg New Energy Finance (BNEF), the renewables-focused analytical arm of Bloomberg, would leave coal with a larger share of power generation than gas in 2040, when it has renewables reaching 50%. That might make sense in the European context on which their forecast seems to be based, but it flies against the US experience of coal losing 18 points of electricity market share since 2007 (from 48.5% to 30.4%), with two-thirds of that drop picked up by gas and one-third by expanding renewables. (See chart below.)

It’s also worth noting that the US Energy Information Administration projected in February that natural gas would continue to gain market share, even in the absence of the EPA’s Clean Power Plan, which is being withdrawn.


Natural gas prices have had a lot to do with the diverging outcomes experienced in Europe and the US, so far. As the shale boom ramped up, average US natural gas spot prices fell from nearly $9 per million BTUs (MMBTU) in 2008 to $3 or less since 2014.  Meanwhile, Europe remains tied to long-term pipeline supplies from Russia and LNG imports from North Africa and elsewhere. Wholesale gas price indexes in Europe reached $7-8 per MMBTU earlier this year.

But it’s not clear that the factors that have kept gas expensive in Europe and protected coal, even as nuclear power was being phased out in Germany, will persist. The US now exports more liquefied natural gas (LNG) than it imports. US LNG exports to Europe may not push out much Russian gas, but along with expanding global LNG capacity they are forcing Gazprom, Russia’s main gas producer and exporter, to become more competitive.

Then there’s the issue of flexibility versus intermittency. Wind and solar power power are not flexible; without batteries or other storage they are at the mercy of daily, seasonal or random variation of sunlight and breezes, and in need of back-up from truly flexible sources. Large-scale hydroelectric capacity, which makes up 75% of today’s global renewable generation and is capable of supplying either 24×7 “baseload” electricity or ramping up and down as needed, has provided much of the back-up for wind and solar in Europe, but is unlikely to grow rapidly in the future.

That means the bulk of the growth in renewables that BNEF sees from now to 2040 must come from extrapolating intermittent wind and solar power from their relatively modest combined 4.5% of the global electricity mix in 2015 to a share larger than coal still holds in the US. The costs of wind and solar technologies have fallen rapidly and are expected to continue to drop, while the integration of these sources into regional power grids at scales up to 20-30% has gone better than many expected. However, without cheap electricity storage on an unprecedented scale, their further market penetration seems likely to encounter increasing headwinds as their share increases.

BNEF may be relying on the same aggressive forecast of falling battery prices that underpinned its recent projection that electric vehicles (EVs) will account for more than half of all new cars by 2040. As the Financial Times noted this week, battery improvements depend on chemistry, not semiconductor electronics. Assuming their costs can continue to fall like those for solar cells looks questionable. Nor is cost–partly a function of temporary government incentives–the only aspect of performance that will determine how well EVs compete with steadily improving conventional cars and hybrids.

I also compared the BNEF gas forecast to the International Energy Agency’s most recent World Energy Outlook, incorporating the national commitments in the Paris climate agreement. The IEA projected that renewables would reach 37% of global power generation by 2040, or roughly half the increase BNEF anticipates. The IEA also saw global gas demand growing by 50%, passing coal by 2040. That’s a very different outcome than the one BNEF expects.

Despite my misgivings about its assumptions and conclusions, the BNEF forecast is a useful scenario for investors and energy companies to consider. With oil prices stuck in low gear and future oil demand highly uncertain, thanks to environmental regulation and electric and autonomous vehicle technologies, many large resource companies have increased their focus on natural gas. Some, like Shell and Total, invested to produce more gas than oil, predicated on gas’s expected role as the lowest-emitting fossil fuel in a decarbonizing world. If that bet turned out to be wrong, many billions of dollars of asset value would be at risk.

However, it’s hard to view that as the likeliest scenario. Consider a simple reality check: As renewable electricity generation grows to mainstream scale, it must displace something. Is that likelier to be relatively inflexible coal generation, with its high emissions of both greenhouse gases and local pollutants, or flexible, lower-emitting natural gas power generation that offers integration synergies with renewables? The US experience so far says that baseload facilities–coal and nuclear–are challenged much more by gas and renewables, than gas-fired power is by renewables plus coal.

The bottom line is that the world gets 80% of the energy we use from oil, gas and coal. Today’s renewable energy technology isn’t up to replacing all of these at the same time, without a much heavier lift from batteries than the latter seem capable of absent a real breakthrough. If the energy transition now underway is indeed being driven by emissions and cleaner air, then it’s coal, not gas, that faces the biggest obstacles.

Original Post

Geoffrey Styles's picture

Thank Geoffrey for the Post!

Energy Central contributors share their experience and insights for the benefit of other Members (like you). Please show them your appreciation by leaving a comment, 'liking' this post, or following this Member.


Joe Deely's picture
Joe Deely on Jul 25, 2017 4:03 pm GMT

Obviously this is a regional discussion. If you are looking at the US there are a variety of different examples.

In SPP – (Southwest Power Pool) renewables are replacing coal. Not much NG in this region.

In ERCOT wind – with a little help from NG- is replacing coal Once coal is gone – wind and solar will be replacing NG.

Out West – coal is mostly gone in CA so renewables are replacing NG. Same goes for Nevada – almost all coal gone so future renewables will replace NG.

Roger Arnold's picture
Roger Arnold on Jul 26, 2017 12:15 am GMT

coal is mostly gone in CA so renewables are replacing NG.

I don’t think that’s correct. Or not quite correct. I suppose one should distinguish between capacity and delivered energy. It’s conceivable that renewables in CA are growing enough to cut into the amount of energy delivered from NG-fired generators. But I’m pretty sure that NG capacity required to support renewables is not declining.

I think what’s actually declining in CA is imports of coal-generated electricity from out of state. CA hasn’t yet implemented energy storage and demand-side regulation at anywhere near the levels that will be needed to reduce the need for NG backing generation.

Willem Post's picture
Willem Post on Jul 26, 2017 12:35 am GMT


Fossil fuels have been about 78% of the world’s energy consumption for the past 10 years. Wind and solar together are about 3%, after about $2.5 trillion invested in wind and solar systems.

The investment in renewables would have to increase from about $300 billion to $1 trillion to move the wind and solar needle at about 1%/y, so the wind and solar % would be about 25% in 2040.

Bloomberg folks usually are optimistic in their estimates.

Robert Hargraves's picture
Robert Hargraves on Jul 26, 2017 10:51 am GMT

Geoffrey, I generally agree. BNEF is over-eager in their projections for wind and solar growth and diminishing battery cost. In the US, regionally, remaining coal power plants are a target for new NG power plants. Worldwide, coal is expanding, because NG is more expensive than in the US, especially after extra costs for LNG. However, NG power is an enabler for wind and solar, because it ramps up/down relatively easily. In my view, the current trajectory for electric power in the US is all NG, with perhaps 20-30% of intermittent chunks displaced by variable, non-dispatchable wind and solar.

Darius Bentvels's picture
Darius Bentvels on Jul 26, 2017 1:37 pm GMT

You forget the major price decreases of wind & solar. Shown in past decades and widely predicted to continue during next decade.

Recent German tender for 1280MW wind, showed that once exceptional expensive offshore wind is now at <3cnt/KWh (incl. decommissioning)…
While further price decreases are expected as bigger wind turbines than the 12-15MW wind turbines on which the price is based, are in development.
Note that those bigger turbines are also higher, increasing capacity factors above the 60% level…

(the 750MW Borssele offshore wind with 8MW turbines will have a CF of 52%)

Mark Heslep's picture
Mark Heslep on Jul 26, 2017 1:54 pm GMT

Since hydro electricity is i) ~3/4 of ‘renewables’ and ii) already developed to maturity in most of the world, then analysis of non-nuclear clean power trends improves clarity by losing the misleading metric of ‘renewables’ and adopting a more accurate description, e.g. intermittent clean power, or ‘non hydro renewables’ to pick up biomass.

In the graph above, breaking out solar and wind would show them at below 10% share in the US and less globally, not about to eclipse either coal or gas within a couple of decades. Then Bloomberg Energy might dispense with such an article.

Hops Gegangen's picture
Hops Gegangen on Jul 26, 2017 1:55 pm GMT

But given the sharp decline in the cost of wind and solar, the next $2.5 trillion buys a lot more than the previous $2.5T. And there are tipping points. Once a source of energy becomes the least expensive, it quickly dominates the market.

And if renewables become cheap enough (and the fuel is free after all) then conversion to methane becomes possible. I was reading recently about a large scale demonstration plant being built to do just that. Then the “natural gas” infrastructure can be used to deliver stored energy.

Helmut Frik's picture
Helmut Frik on Jul 26, 2017 2:13 pm GMT

One point to keep in mind about flexibility: Although coal plants are surely much more polluting than gas powered plants, they will not leave the markte due to inflexibility. Power station engineer – at least here in central europe, have tought those elephants samba and step dancing the recent years.
Or to say it more technical: with the output of solar and wind being well known many hours in advance, hard coal and newer lignite plants (up to 25 years of age) have exactly zero problems to do the neccesary ramping to follow residual load even witch more than 50% renewables in the grid.
So this is not the way to kill coal.
The technology with real ramping problems, prefering negative prices instead of ramping down, are the existing nuclear power plants as market data tells us. Beside must run combined generation which runs according thermal demand.

Joe Deely's picture
Joe Deely on Jul 26, 2017 4:01 pm GMT

Actually both(NG production and capacity) are down but as you say capacity will be much slower to change. Here is link from EIA with latest NG Capacity data for CA – you can see that it is down about 2GW Y-Y. Older simple cycle plants are closing.

Still probably some overkill in NG capacity but you are correct in that much more storage is needed before other plants can close. Let’s see what happens over the next few years.

Coal capacity in the West is definitely in a steep decline. Below is a list of the biggest coal plants in the West and their closure plans.

Joe Deely's picture
Joe Deely on Jul 26, 2017 8:48 pm GMT

Renewables – if you include Hydro – will pass coal in the US around 2025-2027.
In the Western US(CA,OR,WA,AZ,NV,ID,MT,UT,WY,NM,CO) this happened in 2011 and the gap has been widening ever since.

Zero Carbon sources – including nuclear – will pass NG+Coal in the US by 2032.
In the Western US – ZC sources will pass NG+Coal around 2022.

Willem Post's picture
Willem Post on Jul 26, 2017 10:06 pm GMT

Here is the article with the numbers.

Here are some numbers for US offshore wind

Here is the storage Germany needs to have for its wind and solar by 2050

Darius Bentvels's picture
Darius Bentvels on Jul 26, 2017 10:25 pm GMT

Germany has about 30 major pilot plants (~MW-scale) utilizing different technologies and producing different products; H2, methane (CH4), liquid car fuels, etc. Check the lists of projects at the project map link on the right of the page.

They plan to expand the P2G pilots to a total capacity of 2GW in 2022, and start with regular roll-out in 2024.

Sean OM's picture
Sean OM on Jul 27, 2017 7:16 pm GMT

The big punch is with battery storage, you can meet the frequency regulation over capacity requirements without having actual spinning reserves.

Then you add the more flattened grid, so even if demand is higher then generation capacity, you can more easily buy excess capacity from a plant that is already online.

It is actually quite a bit more efficient then how the utilities have traditionally have had to operate. It takes the guesswork out of it, and actually reduces costs for the utilities for variability in either demand or generation.

Geoffrey Styles's picture
Geoffrey Styles on Jul 28, 2017 12:30 pm GMT

The next big increment of investment may buy a lot more capacity, as you note, but the bigger question is whether it will buy a lot more energy (MWh, not MW)? If you consider that the best sites in terms of resource and access to transmission were developed first, then it suggests declining generation as either lesser sites are chosen, or as more new capacity goes into repowering older (better) sites. Looking for a detailed analysis/forecast of this.

Darius Bentvels's picture
Darius Bentvels on Jul 28, 2017 4:09 pm GMT

When you consider the price development of solar and wind in NW-continental Europe (NL, DK, Germany etc.), it’s sure that it will buy a lot more MWh for the buck!

As an example, tendering here in NL concerns the price per MWh during 15yrs, that the bidder needs to construct, maintain, operate and decommission a wind farm of e.g. 750MW+ ~5% positive margin.
After 15yrs the owner has to sell on the whole sale market being on av. ~3cnt/KWh last year.

For offshore wind that price decreased here from >10cnt/KWh towards 7.5cnt/KWh towards 5.4cnt/KWh last autumn (our last tender round). With 0 interest and an av. life of 30yrs that 5.5cnt implies an av. price of 4.2cnt/KWh..

This spring the German offshore wind tender got bids requiring zero guarantee (covering in total 1280MW). Av. whole sale price in Germany was last year 2.9cnt/KWh (it’s somewhat higher this year because many nuclear reactors in France are still stopped because of the certification fraud with nuclear parts).

So in a few years, bidders offer to pay money for the license which allows them to install, operate and decommission an offshore wind farm during 30yrs.
E.g. an yearly fee for that license = right to produce electricity (under certain conditions) on that part of the north-sea …

Sean OM's picture
Sean OM on Jul 31, 2017 10:49 pm GMT

The problem is the fossil fuel plants are not instant start. They need time to get up to speed so they need to be running well before they are needed. What the batteries do, is buy time before you have to start them. In other words, you aren’t running a FF plant unless you actually need the power from it. Right now, they just idle them.

I agree batteries aren’t a cost effective solution for every problem right now, but as they drop in price, they are becoming more cost effective in more places. In 10-20 years, they may be the de facto standard, but there is a lot of work to get done before we get to that point.

Get Published - Build a Following

The Energy Central Power Industry Network is based on one core idea - power industry professionals helping each other and advancing the industry by sharing and learning from each other.

If you have an experience or insight to share or have learned something from a conference or seminar, your peers and colleagues on Energy Central want to hear about it. It's also easy to share a link to an article you've liked or an industry resource that you think would be helpful.

                 Learn more about posting on Energy Central »