>Thanks Naomi, I figured if it could be done I would get the help here. There is one additional quirk that I failed to mention. I need to select a specific AmendmentNumber from both tables. So in my real world case I ended up with this:
>

>;with cte as (
>select P.AmendmentNumber, coalesce(P.TipId, A.TipId) as TipId, coalesce(P.Amount,0) - coalesce(A.Amount,0) as Amount,
>coalesce(P.FundType, A.FundType) as FundType
>from TIP_AmendmentProposedFunding P 
>full join TIP_AmendmentApprovedFunding A 
>on P.TipID = A.TipID and P.FundType = A.FundType
>where P.AmendmentNumber = 2014.01 and A.AmendmentNumber = 2014.01
>)
>
>select * from CTE PIVOT (sum(Amount) for FundType IN ([ASTP], [RTA], [HURF26], [DIFO], [Local], [STP])) pvt
>

>This gets pretty close but the computation is off. here is some real data.
>
>APPROVED
>Keyid AmendmentNumber TipID AMOUNT FundType YEAR PHASE
>296 2014.01 5.12 8000 NH 2013 DR
>297 2014.01 5.12 16500 RTA 2017 D
>298 2014.01 5.12 6500 ASTP 2017 D
>
>PROPOSED
>Keyid AmendmentNumber TipID AMOUNT FundType YEAR PHASE
>332 2014.01 5.12 8000 NH 2013 DR
>333 2014.01 5.12 16500 RTA 2017 D
>334 2014.01 5.12 6500 ASTP 2017 D
>336 2014.01 5.12 42900 RTA 2018 C
>337 2014.01 5.12 43583 ASTP 2018 C
>
>For this project, the resultant difference for ASTP is 37083, whereas it should be 43583 because the two 6500 values cancel each other out. So maybe there needs to be a couple of sums and group by in the first part of the select? It is still fairly greek to me.
>
>Its interesting that the keyword PIVOT does not appear in my 800 page SQL handbook. Neither does pvt which I assume is just an alias assigned to the result. Correct?

You're right, looks like you first need to group by FundType and then join.

PIVOT was added in SQL 2005. Prior to that (in SQL 2000) you would use case based pivot instead.