The frequent flyer miles are STILL a non-issue. That someone pays for the ticket on a card that generates miles versus one that doesn't is a choice. Such miles these days almost ALWAYS are generated for the individual, since changes in security mean that individuals must purchase tickets with their names on them, not just a ticket for XYZ company employee. Frequent flyer miles are NOT a part of a budget (they really CAN'T be) - ergo, they inure to the benefit of the individual (and the individual's credit card that is being used for the convenience of the employer AND the grant funder). Miles belong to INDIVIDUALS. If you're eating at a restaurant, and the meal is paid for by the employer or grantor, and the restaurant gives you a coupon to come back for a free meal - you don't owe the employer or the grantor that amount. You got a coupon. Enjoy it.

There is NO federal funding program and I've ever seen (and yes, I've actually got a good number of federal grants under my belt for a variety of organization types) where the per diem amount that a grant will pay comes close to the actual costs for a single individual, much less TWO people. If there is reimbursement occurring that is covering ALL food/beverage expenses, I'll guarantee you that the employer is covering some of that expense. The feds just aren't that generous and make MINIMAL amounts available for meal expenses - amounts that are usually set at a level across the board, no matter where the attendee is eating.

If there is evidence of fraud such that the expenses are being claimed in the course of attending conferences that are not actually being attended, I'd agree with you 100% that it's fraudulent. Again, however, one would need to see the guidelines to determine the requirements of the funding. Must they attend EVERY session of the conference? Show up for one, two, six? Some grants are specific, others are very open ended. For conferences where things like licensure or a certain skill set need to be maintained, those conferences often have a system such as punch cards, certificates, or other mechanism such that only those persons who actually attended any given session are given credit for that session. If your butt isn't in the chair for the lecture, you don't get credit for it. For scientific research symposiums, the standard doesn't apply - they are more of an exchange of ideas, thoughts on directions that research might be taking, etc. - the difference being that such symposiums presume that attendees have their own vested interest in attending for their own benefit, rather than attending because they have to have X number of credits to turn in to a licensing agency. As they have no requirement to submit proof of attendance, yes, such symposiums probably have a high rate of "cheating".

You can certainly address your concerns to the employer, who initially approves such expenses. But it would behoove you to read a few successfully funded proposals and see their budgets before boiling over. Conferences CAN be fun. People CAN bring people with them. They tend to be held in exotic locations for a REASON - to GET people to attend. Notice how you never see conferences in Hell, Montana? That's because people won't GO THERE. Federal funding will generally pay for rooms and expenses for the term of the conference, plus the day before, and the day after, to accomodate travel. So for a 1.5 day conference, yes, 3.5 days is not only acceptable, and reasonable, but actually EXPECTED.

In the BIG picture of the issue, this isn't going to be about the expenses. It's just not. Receipts have been submitted and approved. You are barking up a tree where nothing is going to fall. If there is an angle of concern here, it is the non-attendance issue. Go at it from that angle.