How hard is it to set up a Vancouver startup independently, if you have 2-3 US citizens who want to relocate? Assume you have $100-200k of your own funding.
Gaming the tax credits seems like it takes a bit of effort, but lowers the effective tax rate substantially below even Nevada.
Starting a company in Canada among 2 or 3 people is easy. I should know because I did it when I was a teenager. It cost me $15 for some incorporation templates, $150 for a name search report showing potentially similar company names to the one I wanted to start, and $350 or so in government fees. This got me a company incorporated provincially. You can do it federally too but it's a bit more complex. If you have as much money as you claim you could probably emigrate to Canada in a special business class meant specifically for people starting businesses here.
Starting businesses is easy as pie. If you have enough money to sustain the business before it brings in revenue then I'm not sure what the attraction is to these sort of ventures. If your execution is good then VCs will hear you out regardless of whether you've been "accepted" into a program like this. They may facilitate introductions but there are also government funded programs here in Canada that do not charge money which will do the same.
Tax rates for people smart enough to start businesses in Canada are fairly attractive. It is especially so when you factor in healthcare costs or lack-there-of. However one thing to bear in mind is that it is often a severe hassle an American company to exit an investment in a Canadian company. In some cases it would require 900 signatures of limited partners in the American company and for each to file Canadian income taxes. Google "Section 116 foreign investment". It is something the government here is tackling in a bid to spur VC investment in Canada. But there are still some gotchas (and benefits) to starting a company here.
I like accelerators for two reasons:
1) Having a same-time-founded cohort of companies. Moral support, entertainment, and when ~80% of them fail, you can pick and choose assets (great developers) having great visibility into them.
2) I like the idea of accelerators for people with $0 and just out of college or high school, or who need structure. If I could give 5% of my company to give an accelerator a win (both in money upside and reputation), I'd do it, especially if I were in a third-tier city like Vancouver which deserves a second tier startup population.
I think the odds of failure at 50-80%, including my own company, and having a bunch of ready ways to re-deploy those people into other projects is good for EVERYONE -- the engineers can move to the most successful projects. If I worked on something for a year or two and it flopped, I'd love to go work at a startup I'd been following the whole time, rather than trying to find a regular job.
I think SFBA in general is basically an incubator/accelerator, if you've been involved in one startup even as an engineer. Formal accelerators help a lot more in places where most of the people who want to do a startup have NOT yet worked for one. College would be the best example.
Gaming the tax credits seems like it takes a bit of effort, but lowers the effective tax rate substantially below even Nevada.