Of course, it depends what you're familiar with. I could assume unfimilarity with SQL Server, when a concept like staging table is unknown to gussy81, because it's in general a way to first just import what's in the origin 1:1 and then use your T-SQL datatype knowledge, knowledge about CAST and CONVERT and other T-SQL functions like I used, to convert after input.
Maybe I'm too fast in my judgment thinking lack of T-SQL knowledge means lack of other programming knowledge, too. I think the batallion of functionalities you have in T-SQL is always fitting any needs, even unconventional (at least in SQL terms) numeric formats.
Regarding dates the CONVERT function shows MSSQL knows many standards it can convert from a character represnetation to one of it's date/time/datetime data types. There's nothing so specialized about numbers as they are also usually not formatted in such a unusual way. And even with an extension of the CONVERT function capabilities - quite feasable, if you think that Microsoft wants to support many standards to make imports easy - it likely won't make it into implicit conversion without explicitly calling the CONVERT() function, i.e. it's not feasable to interpret a string like '100-' as -100, even if you know about accounting debt notation. Excel obviously does that. I haven't tried it out. So you could argue software of MS isn't consistent. Well, always keep in mind MS did not develop anything it now has in mainly Windows, Office applications and SQL Server from scratch by itself with a consistent code base across all applications and servers. IIRC the Office applications were even put together from different acquisations, so just because Excel makes 100- to -100, doesn't mean every MS software does.
So, last not least, the problem could likely also be solved in the output of such data to the much more conventional sign before number format, especially when the output is to CSV, CSV is not an accounting standard. It's a standard lacking some standardization, and as any combination of characters can always be a string, there's nothing that would trigger a warning or error if you output ...,100-,... into a CSV file. There's usually not even an association of CSV with database software, I thikn the default file association is notepad or Execl, once you install Office, but eve n with such an association a software outputtting a file with a specific file extensions doesn't pull up an instance of associated software to be a witness and verificator of the correctness of the file content. So, in short, while it's possible to finally have sofware like Excel importing 100- as -100, you expect too much of such a standardization and you better convert to norms that are more widespread, if you output data for transfer to other systems.
So, that is a task that's well justified to be a burden for the output of such data to exchange files than it is your task to handle it in input. Of course, there's no point in demanding the data exchange format, but you can argue for a better exchange file, on the basis it surely is less work to change the export than your import. One thing I can assure, when they use their own exports to import them in Excel: Excel will import -100 as -100 , too.
Chriss