4 ms·
Several decades of politics and industry kicking the can down the road and hoping that the challenges go away by itself. For example: German car makers are sti
by pgeorgi 3y ago
Several decades of politics and industry kicking the can down the road and hoping that the challenges go away by itself.
For example: German car makers are still somewhat reluctant to embrace electric motors (no matter the power source or storage) and would like to stay with combustion engines (no matter the combustible) because that's a niche they and the entire downstream industry feel comfortable with.
Yes, a large share of the industrial base in Germany hinges on it. But a long-term view on things would have started divesting from this exclusivity decades ago. Instead, they hoped the challenge just goes away by building better combustion engines. There's _finally_ some movement in that direction (Volkswagen embracing EVs is huge), but there are also still reactionary pockets that try to water everything down because hydrogen or synthetic fuels are right around the corner, just you wait! (possible, but very unlikely)
Another example from the industry: The chemical industry never hedged their reliance on fossils, either. While they will remain a consumer of fossils for chemical processes, the industry was comfortable just heating whatever they need to be really hot with them as well. Now they're scrambling to retool their factories to still use fossils where impossible to replace (e.g. fertilizer production uses the chemical properties) but to use hydrogen when stuff just needs to get hot. That the amount of fossil fuel might go down was _kinda_ obvious with all those CO2-reduction targets, but they preferred to wait it out.
I see more hope with them, though: they need to retool, but after the furnaces use hydrogen (wherever that comes from), they can do business as usual (and gradually optimize hydrogen sourcing). The car industry (some say that 20% of the German industrial base rely on that directly or indirectly) is much worse off.
Another example, this time politics: Deutsche Bahn (DB) is a private company fully owned by the government. There's some money going back and forth between the company and its owner, but one of the arrangements was that DB pays for maintenance of its infrastructure itself, but catastrophic loss of infrastructure (and related repairs/rebuilds) are covered through government money. What is the private company (that was expected to pay some 8-9 digit sums of earnings back to the federal budget for years as part of that back-and-forth) doing then? Cut down on maintenance. When stuff breaks entirely, repairs aren't on them, so that's the "rational" choice for the company. Politicians were asleep at the wheel as far as these incentives are concerned (or worse: they calculated with them, kicking the can down the road.)
There are advocates for moving more logistics from the road to rail, but that's kinda stuck because there's a huge repair-and-investment plan right now (to deal with that oversight of decades) that will block various key routes for the next few years.
And it's not as if the Autobahn-network is any better off (map with construction sites at https://www.autobahn.de/die-autobahn https://www.autobahn.de/die-autobahn): lots of bridges from the 60s reaching their end-of-life all at once because maintenance is an expense that was better ignored (since politicians decided that above all else, the country mustn't increase its debt: So we had a kinda-balanced budget but accrued the steel-and-concrete equivalent of tech debt)
- pgeorgi 3y agohttps://dserver.bundestag.de/btd/20/102/2010284.pdf https://dserver.bundestag.de/btd/20/102/2010284.pdf has a list of railway infrastructure elements (track switches, bridges, overhead electrical lines, rail tracks, track crossings, signal boxes, support structures) that are, while still safely usable, in the condition categories "einschränkend" (limiting), "schlecht" (bad), and "mangelhaft" (deficient) - that is, the condition categories that require some kind of maintenance and might reduce efficiency of the rail network. Starts on page 9, ends on page 494.