I suspect that the idea that merged cells represent a great evil, though representing received developer wisdom, depends very much on one's use case. After all, only a small proportion of sheets contain formulas (<5%) whereas over 50% contain merged cells.
My personal acceptance of merged cells (in particular to create form-like interfaces) stems from the fact that I do not employ the usual interactive practices of spreadsheet development which are built upon manipulation of the grid. For me, the sheet is simply a blank canvas and it is the objects (tables, named ranges, array formulas, pivot tables, charts … ) that I address by name. Their actual location is an irrelevancy determined by journalistic and graphic design considerations.
For coding purposes, I believe the address of a merged region is the address of its top-left cell but, then again, I never use cell coordinates to addresses the worksheet from VBA; I find range names and table names far more reliable.
The VBA required to convert a formula to its value is the somewhat odd looking line of code:
Code:
Sub ConverToValue()
[mergedCellName].Value = [mergedCellName].Value
End Sub
This works for a merged region just as it does for a single cell.