This could become expensive for citizens in the long term

WirtschaftsWoche: Professor Rosenow, the heating debate in Germany just won’t die down. The federal government wants to revise the Building Energy Act in January. At the same time, the first major cities are currently presenting their municipal heat plans. What stands out is that gas heating is also expected to play a central role in the future. The networks are to be maintained, even expanded, and gas boilers are later to be operated with green hydrogen.

Jan Rosenow: That way, the cities will have formally fulfilled their legal obligation to present a municipal heat plan. But by placing such a strong focus on gas heating, they are imposing immense risks on their citizens.

What risks?

I see a relatively large danger that such plans may be convenient in the short term, but in the long term will prove unsustainable or barely affordable. In any case, they risk becoming very expensive for citizens.

How so? The gas network already exists and is intact. There are now several feasibility studies showing that it can be converted to green hydrogen with manageable effort. That would make gas heating climate-neutral. What bothers you?

First of all, one has to look more closely at who calculated what, and with which specific assumptions. Unfortunately, some studies are not neutral, but are commissioned by gas network operators.

What interest do gas network operators have in “tuned” numbers?

The German gas network is a huge asset, grown over decades to more than 600,000 kilometers in length and branched into even the smallest corners of residential neighborhoods. Naturally, the operators want to preserve it for as long as possible. Every single gas customer who switches to an alternative — whether a gas tank, biomass, or a heat pump — is probably gone for good and reduces the value of the asset and its potential to generate returns in the future.

Individual crises such as the war in Ukraine and the loss of cheap Russian gas supplies can be weathered, because higher procurement prices for gas can ultimately be passed on to customers. But a shrinking network — and thus a shrinking capital base — is, in the long run, the inevitable death of any business model. That may partly explain the intensity and emotionality with which the debate about building heating systems is conducted.

But hydrogen would then be the perfect solution for both sides: customers wouldn’t need an expensive heat pump, and the existing gas network could continue to be used.

Technically, the gas network can be converted to hydrogen. But the effort involved is not as small as it is often portrayed. The advantage of the idea is that it can be implemented gradually. Climate-friendly hydrogen can initially be blended into natural gas in small quantities; the existing technology can handle that, so customers don’t notice anything at first. But blending doesn’t achieve much in climate terms, because hydrogen has a much lower energy density than natural gas — only about a third. So even if you replaced 30 percent of the fossil natural gas coming out of the pipeline with climate-friendly hydrogen — which we are still miles away from — you would only have replaced about ten percent of the energy content, and thus only ten percent of the CO₂ emissions.

What does that cost?

At present, costs are around ten euros per kilogram, or 30 cents per kilowatt-hour. Even if you calculate very optimistically for the future, by 2035 you still end up, on average, with at least double the cost per kilowatt-hour compared with today’s natural gas. And that’s just production. Transport and storage costs of a similar magnitude are added on top. Saying the natural gas network is “H₂-ready,” meaning suitable for hydrogen, is a bit like a low-income earner saying, “My parking space is Ferrari-ready.”

Okay, but the production of green hydrogen is still in its infancy. With mass production, costs should fall.

To be honest, I’m not that optimistic. For one thing, we’ve been promised that for ten years now, but the quantities produced are still very small. We see not the slightest evidence of scaling up volumes and falling costs. On the contrary: according to the very latest figures, costs are actually rising again. At this point, people like to argue that surplus wind and solar power could be used for hydrogen production, which would cost nothing because it currently has to be curtailed. But on closer inspection, there isn’t that much of it. In any case, surplus wind power is nowhere near sufficient for the quantities of hydrogen that would be needed for use in gas heating, and surplus solar power doesn’t occur during the heating season, which would entail immense storage costs.

Hydrogen could also be imported.

But that would require a completely different pace of ramp-up in the countries of origin. Unfortunately, that isn’t happening either. The problem with hydrogen is simply its complex production. Even if electrolysis were to scale as fast as the fastest-growing technology in recent human history — which was warplanes and warships in the United States during World War II — by 2040 at most only eight percent of today’s fossil-fuel-supplied sectors would have been converted to hydrogen. And that is pure utopia; the hydrogen industry is light-years away from the pace of U.S. aircraft construction in the 1940s.

Some politicians and energy utilities argue that not all, but only some of today’s gas customers would have to switch to hydrogen — for example, those whose poorly insulated houses make heat pumps inefficient. Wouldn’t it also have advantages if not everything had to be electrified?

For one thing, the very high hydrogen costs argue against it. Who is supposed to pay for that? Above all, there are also the network charges: if only part of today’s gas customers remain connected to the network and the rest switch to alternatives such as heat pumps, the network charges are spread over fewer and fewer shoulders. In the end, they rise exponentially. There is no satisfactory solution to this problem: either the last gas network customers get bitten by the dogs, or the legislator decides on a bailout, in which case we all pay the difference as taxpayers — and for a very long time, because a new gas boiler installed in 2026 will run until 2046 or longer. From an economic point of view, it would be cheaper to help the affected gas network operators financially to dismantle their networks than to heavily subsidize millions of end customers for decades.

So what should be done?

If we ask only engineers and physicists, the answer is crystal clear: central district heating networks for large apartment buildings and commercial buildings in densely built inner cities, and heat pumps in detached houses in outlying areas. A heat pump is simply six times as efficient as a gas boiler with hydrogen, if you correctly calculate the entire chain from fuel production to the heat that arrives at the radiator.

Many homeowners are afraid of high renovation and electricity costs when switching to a heat pump. Many houses are in poor energy condition.

I know that very high figures are often cited in Germany. But the empirical evidence doesn’t support that.

The city of Chemnitz, for example, recently officially estimated renovation costs of up to 200,000 euros per house in its municipal heat plan for switching to heat pumps.

You can only arrive at 200,000 euros in energy renovation costs for an average single-family house if you assume converting the central heating system from radiators to underfloor heating and a complete external insulation system, plus new windows and possibly better roof insulation. According to empirical data, such an extensive renovation is not necessary in 90 percent of existing buildings for efficient operation with a heat pump.

The Fraunhofer Institute, for example, has now measured and analyzed hundreds of existing buildings after conversion to heat pumps. We have similar data from England, Sweden, Austria, and the Netherlands — countries where houses are, on average, no better in energy terms than German ones. Realistically, some undersized radiators need to be replaced; that costs 300 to 400 euros per unit. Some windows, if they are very old, too — but that would be advisable with any type of heating. Insulating the top floor ceiling and the basement ceiling are also measures that cost little and have a relatively large effect. I myself live in a Victorian house from 1880. It has been kept comfortably warm every winter with a heat pump for six years now; my heating costs are lower than they were with gas.

Because electricity is cheaper in England than in Germany.

Gas is cheaper there too.

Okay. For a nationwide switch to heat pumps, the electricity grids would have to be significantly expanded. We’re talking about hundreds of billions of euros over the coming decades.

That’s true. If you electrify a sector that has so far been largely fossil-fuel-based, like building heating, you need larger power grids. But part of the costs are due to digitizing the grids and expanding them for other applications, such as further electrification of industry, data centers, and electric vehicles.

But a significant share of grid expansion costs is already caused by heat pumps, because their load cannot be spread over time as well as, for example, EV charging. Peak loads from heat pumps occur pretty much simultaneously: when it’s cold.

That’s correct. However, people often calculate with extreme values here. If I design the distribution grid for the peak load of 15 million additional heat pumps on a cold winter evening, it naturally becomes expensive. But we do have some flexibility in the load from heat pumps: there are thermal buffer storage systems that can bridge several hours. That already helps quite a lot with electricity prices and grid bottlenecks. Large batteries on the electricity side, in the grids, which smooth out price spikes, are developing rapidly, both technically and economically. They are becoming cheaper and are therefore increasingly designed for several hours of storage. Of course, grid expansion still costs a lot of money. But with increasing electrification of industry, building heating, and transport, the grid charges will then be spread over many more kilowatt-hours.

So the effect is the opposite of the gas network, where fewer and fewer customers have to bear ever higher network charges per kilowatt-hour of gas?

Yes. The political problem is simply this: the costs of expanding the electricity grids arise mainly at the beginning of the transformation; the revenues are long-term, through avoided oil and gas imports and avoided CO₂ costs. For politicians and municipal utilities — who are often also the operators of gas networks — it is naturally very tempting to promote a pseudo-solution like hydrogen in gas heating. It costs little at first, but the big bill comes at the very end.

This article is also published on Medium. illuminem Voices is a democratic space presenting the thoughts and opinions of leading Sustainability & Energy writers, their opinions do not necessarily represent those of illuminem.

Track the real‑world impact behind the sustainability headlines. illuminem’s Data Hub™ offers transparent performance data and climate targets of companies driving the transition.