Forgive me my ignorance, but: parts from which cars are assembled (or raw materials from which parts are manufactured), are also subject to tariffs, aren't they? So the only shift that would happen is that of the labour (and US labour is not the cheapest, IIRC).
They would in the current scenario, yes, but the OP said “Two typical scenarios that we know from the past in industries like cars for example”
In the past, countries would put tariffs on importing cars, but not on importing car parts (with some complex definition of what constitutes a car and a car part. IIRC, there once was a loophole where one imported a car and converted it into a van by removing back seats to avoid a tax on importing vans)
For manufacturing physical goods, labour cost is a small percentage of the total cost of the good. Why is this? Because modern labourers are extremely productive: they are highly skilled at their jobs and use very efficient tools and machines to do their jobs.
The problem with this theory is the tariffs will both have to remain in place for a decade and businesses will have to believe that they'll remain in place for a decade for that to pan out.
Nobody is building a new factory, based on tariffs that they expect to be gone in three years time.
They will only shift if it's predictable that the tariff policy will stay for long enough to recoup the high capital investments of building a whole new factory just to serve the internal market.
For many industries it might not make sense unless it's a 5-10 years long plan, the risk of investing a lot to build a new factory, bringing it online over the next 2-3 years, to then have tariffs removed and making your new shiny factory more expensive to run than one outside of the country, will also be factored into the total cost.
The tariffs are so broadly applied that the risk factor is much more massive than anything the USA experienced before (like during the Japanese cars era of the 70s/80s), it's wishful thinking the domino effect will happen in the short/medium-term.
