And that argument is almost always silly. All of these big companies are devoting resources all the time to reducing payroll. It is banged into their heads in MBA schools that "payroll is your biggest expense", and that is a bad thing. With that mindset they are always looking to reduce payroll.
Yes, this strawman you have created is a silly argument. The size of the payroll has nothing to do with it.
The decision to automate or not is decided by profitability. If a worker's wages is less than the cost of automation, then worker is not automated. But once wages exceed the cost of automation the worker will be replaced. So, increasing wages will lead to more automation.
This is such a smug take, that the warehouse workers should not anger the big almighty Amazon, or they'll automate their job.
If Amazon need them today, they have their value. This gives the workers even more incentive to squeeze Amazon ASAP, while Amazon is desperate for manpower.
You could just as well argue that increased labor costs due to unions means Amazon will be forced to spend less on developing automation, since they don't have the budget for it.
Minor increases wouldn't be worth organizing an union over, would they?
Or they are on the order of 30-40 per cent, which no longer falls under "minor" and can turn some kind of automation technology barely, but profitable.
Not everything is matter of "many people mashing the automation buttons". Business involves a lot of trade-offs and the decision making is pretty complicated. Neither you nor me can really predict all the butterfly effects involved.
I wish them success, but presenting the wage/budget situation as totally immune to balance changes is unrealistic.