Is there a quantitative analysis somewhere of who is effected by government shutdown and by how much?
Beyond that you have all the second order effects like landlords and lenders not getting paid on time which starts to become a problem around 2-3 weeks in when the first pay day is missed.
EDIT: Wow, couldn't type well on my phone this morning.
For the workers who are actually sent home and correspondingly never get paid, that means the government still has the money it would have had to pay them. So either it goes back to the taxpayer (in theory) or more likely they find some other way to spend it. Either way it goes back into the economy again.
So it's not about the money.
There are some other issues though. One is you lose whatever value the government employees were providing. Though this is mostly the broken window fallacy. If you give someone a month off, they may still do something productive with it -- work a temporary job, catch up on home repairs, take a vacation or attend a conference etc. Whether or how much this is less or more productive overall than their government work is a case by case determination, but it's not as if typical government work is renowned for its efficiency.
Another issue is that if it extends for too long, the furloughed employees will find other work permanently. That can be a problem if you only have to turn around and pay recruitment and training costs to replace them again. On the other hand, it may actually help to thin the herd -- it's a lot easier to reduce or eliminate inefficient programs after some or all of the people who normally lobby to keep their jobs are already gone.
Because the landlord needs someone in the home, he may lose the opportunity. A damned shame.