Enter "A Financial Analysis of Consumer Mortgage Decisions," a new report by the Research Institute for Housing America and the Mortgage Bankers Association. It discusses various mortgage choices and when it makes sense to get what, including a refinancing.

The authors' advice: Don't just consider interest rates. Look sharp at the closing costs.

Play around with this "Optimum Mortgage Refinancing Calculator," noted in the report, and you'll see how cosing costs can make a difference.

Say you've got a 6 percent interest rate on a mortgage with a balance of $250,000. You're contemplating a refi into a loan that has a 5.5 percent interest rate and 1 discount point.

If your closing costs add up to $2,200 and you roll that into your new mortgage along with the point, you'll save $137 a month over your old loan.

But what if you have $5,000 in closing costs, at the higher end of what LendingTree says the range can be?

Then your monthly savings are $121. That's $192 less a year than the savings with the cheaper closing costs.

Ah, you say, but a savings is a savings. Who cares about the closing costs if you can roll them into the mortgage amount and still pay $121 less a month?

Because, the authors say, you want to think about the "opportunity cost" of refinancing now only to see rates fall further. If you refinance now, it might not be in your best interest to refinance later as well.

"If closing costs are significant, say 3 percent of the remaining principal, and you refinance every time there is a 50 basis point drop in rates, you may never fully recover the cumulative closing expense incurred," they write. (And meanwhile your principal keeps growing. That's why the "serial refinancing" encouraged by some -- ah -- "helpful" folks in the mortgage industry isn't a good idea for a typical borrower.)

The mortgage calculator I linked to above attempts to take all this into account and tell you whether -- given your potential interest rate and costs -- it's a great, OK or bad idea to refinance. Its opinion on the refi opportunity with $5,000 in closing costs: Pass. It dubs the one with $2,200 in closing costs "OK, But Not Optimal." (What would be optimal in this case? Closing costs under $1,250. Or $2,200 in closing costs with a 5.4 percent interest rate rather than 5.5 percent.)

The authors dub this "refinancing efficiency."

What do you think of this way of looking at refinancing options?

And something to consider in today's environment: Does it make sense to hold off refinancing on the chance that rates could go lower when the prevailing expectation is for increases?