The biggest problem I have with the OMIGOD! deficit debate is that everybody thinks that the solution is to throw all the Federal programs into the mix. So, even though Social Security is fiscally sound, has always operated under its own budget, and is good for the near term, "reasonable" folks want to allow it to be gutted.
But many of us know that the way you deal with very large seemingly intractable problems is to split the problems into smaller pieces, and to then deal with those smaller problem individually. For Social Security (and Medicare) this has already been done. Medicare is certainly a significant problem with an estimated $5trillion deficit over the next ten years, and deserves focused consideration.
The rest of the near-term deficit difficulty comes from the "U. S. Government" budget, which currently consists of interest payments, "mandatory" spending, and (non-)security spending. If "divide and conquer" is the way to go, should we consider chopping the "U. S. Government" budget into virtual pieces, giving (e.g.) Defense its own revenue stream (e.g. a percentage of tax revenue)?
It seems that India decided (some time ago) to split its rail budget from the rest of the national budget because spending on rail was so great. Maybe that's the right thing to do when fiscal times are tight. This is also what happens in the corporate world when a large amorphous company is having problems: the CEO tasks each division head to get that division into the black.
The current U. S. Government budget gloms all spending (over the next ten years) into four buckets:
Security........ $9187 billion
Non-security.... $5120 billion
Mandatory....... $5532 billion
Interest........ $6029 billion
So, what are we really talking about here? My guess is that we start with the existing "U. S. Government" budget and assign each Department to a bucket (e.g. DoD, Veterans Affairs, and Homeland Security go into the "Security" bucket). Then there has to be some inter-bucket budgeting. For example, the Department of Energy takes care of all the plutonium used for weaponry. If DoE is in the "Non-security" bucket, then there should be a transfer from "Security" to pay for that.
Finally, allot a percentage of each revenue stream (not currently devoted to other causes) to each bucket, and see if you can get each bucket's budget in (reasonable) balance.
For example, let's say that the projected "Security" spending for the next ten years is $9 trillion (above) plus another $1 trillion (payments to DoE and NASA, e.g.) plus another $3 trillion (as a contribution to the "Interest"). That's $13 trillion.
Now, let's say that we're willing to put 50% of our personal income taxes into "Security" plus 75% of all corporate taxes, plus all the estate taxes, excise taxes, and duties. That's about $13 trillion over the next ten years. Terrific!
Sort of sucks for everyone else, though.
And I don't really want half my income taxes going to "security".