Thank god Switzerland has the 2nd pillar, else we would have a gigantic catastrophe eventually. Like Germany will soon have.
At least 2nd pillar secures a bare minumum for most.
1 Like
Doesnât Germany have a very similar pillar system? I know Netherlands has, and pretty sure the UK has as well. Itâs not too bad if someone doesnât want to RE I think.
Unless people empty it to buy over priced real estate 
8 Likes
So, back on track? Nothingburger? Sky is not falling anymore?
1 Like
Germany only has an equivalent of AHV, and with a lot worse demographic problem coming.
Some very select companies have their (worse) version of 2nd pillar/Betriebsrente. But thatâs really nothing meaningful on population level.
Nothing 3a like at all. (there is only Riesterente which is basically a legalized insurance scam)
Itâs basically just worse AHV for the average Joe there.
2 Likes
TY, didnât know, my German colleagues told me our company (our employer) whoâre basically a SME does have âprivate pension with employer matchâ.
That said, we had the same in Greece (same employer) with an insurance company (anyone can get this, employed or not), and our employer added 100% on top of our contributions, thatâs not that uncommon in Greece, many companies do it, not sure about DE. I guess the difference is that in CH this is mandated (which is a good thing).
1 Like
- Donât use pillar 3a â 14k available p.a. (as a couple)
- Car leasing (half of all cars in CH!) â 18k p.a. for a 100k car like Audi S6
Lots of people here could live such a lifestyle if they wanted.
1 Like
Did anyone notice SMI performance for 2025 so far? Seems like itâs performing better than everyone else
One month of glory at least
5 Likes
Yep, no need for S6 either, S4 or BMW M3 or these beautiful beastly Dodge GTs are more appropriate for 20-somethings with good jobs, I see them in my garage and tell my wife âsee, no kids, ah the dreams that were shatteredâŠâ 
2 Likes
I think SMI is not even at all time high yet. So should be fine. Of course the challenges are with Pharma and what US policies will do
âPharma companies donât treat US fairly and overchargeâ is the rhetoric these days
Seems that Deepseek discussion with all its repercussions has already been put to rest. Hereâs the benchmark comparison with other LLMs in case someoneâs still interested:
1 Like
Iâve been using DeepSeek for a long time now and the model is good, but not earth shattering. The main impacts are:
- Efficiency gains and technical innovations
- Low cost
The efficiency gains were a necessity given the lack of access to newest GPUs. With only access to H800s which have certain limitations, they had to get creative to work around it. The whole architecture is a masterclass in a unified design to work around the limitations they had and once again proves that necessity is the mother of invention.
The low cost will have implications. While the best models will always be able to charge a premium, when you have something that is, say, 80% as good as a top model while costing 5%, then this damages the revenue potential of the likes of OpenAI and Anthropic.
6 Likes
So basically more or less on par with existing models and free of charge. I guess happy times for AI users as AI gets democratic
No need for professional hardware. AI (deepseek, llama) can now be run locally on a laptop. Very moustachian 
These laptop compatible versions are distilled versions of Deepseek r1 which are much weaker.
2 Likes
Too many perverse incentives in the US healthcare system. A Yank once told me âThe US pays for Europeâs treatmentâ, this sentence is so dumb that I had to respond with ââŠand defence, arenât we smart?â.
Fidelityâs Jurian Timmer:
The resilience in corporate earnings goes hand in hand with rising growth estimates for the US economy. Since the soft-landing scare of 2022/2023, GDP estimates have been on the rise and are now close to 3% (real). Add 3% inflation and you get 6% nominal GDP. A 10-year government bond yield of 4.6% doesnât make much sense against that backdrop, especially without the help of an accommodative Fed.
(Source)
4 Likes
I think, (relative) log charts would help. There is just more relevant information in them. The orange and the blue lines look the same, but the orange line grew about 2x as much over the last few years. Of course, after some years of strong underperformance. And a deceptively similar COVID drop (in reality also ~2x).
1 Like