Number of people being irrelevant might mean that a welfare recipient doesn't work but enables someone else to work, or if someone doesn't get welfare, they work, but their support is withdrawn from someone else. It could also mean that the benefit is not enough for them to consider not working. That is to say that their ambitions are greater than merely surviving, the enabling of which is the foundational goal of a reasonable welfare program.
I think I would agree that welfare is a corporate subsidy in some sense, but it covers more people than corporations would without the subsidy, including people unable to work or for whom paid work is less beneficial than some uncompensated task.
Another point to raise is that in poorer parts of the world there is far less, proportionally, that is provided by corporations, large or small.
If they've determined that welfare has statistically irrelevant impacts on the # of people participating in the labor force and the # of hours worked that would seem to imply that welfare recipients are still a needed part of the workforce. That would also lead one to believe that welfare is not allowing substantial lifestyle modification or opportunity for self-improvement.