Your numbers may be high. Reports say 800 jobs, hourly employees don’t work 40 hours every week of the year, etc.
But let’s take it as a high-end estimate. There are some local businesses that would make less without the Carrier employees as customers, which you didn’t include in your calculations, so that could make up the difference.
The problem is that you assume none of the people who lose a job at Carrier would find work elsewhere. If that’s true, then they’ll get unemployment benefits, Medicaid, and food stamps, costing the government. However, if they do get another job, then they’ll pay taxes.
The net effect on government finances therefore depends on a number of unknowns, but probably would be lower than your estimate.
Regardless, the reason the Carrier deal is bad economics is the effect on incentives, not tax revenue. As I noted in the article, it encourages every company to seek a bribe by threatening to move jobs to another country. And it determines economic outcomes by connections and political usefulness, rather than business competitiveness, which is highly inefficient.