PostgreSQL, don't leave home without it... MongoDB <giggles>, I guess all the smart kids would know to prejoin their data... MySQL <giggles>, at least it could do half of it...

rocknelephant.sql

COPY(
SELECT 
  p.id as "id",
  p.payor_email as "email",
  to_char(p.created_at,'YYYYMMDD') as "date",
  pgroup.gross*0.01::money as "gross",
  pgroup.refunded*0.01::money as "refunded",
  f.name as "fund",
  pgroup.fee*0.01::money as "fee",
  (SELECT 'Paid') as "status",
  p.last4 as "last4",
  p.payor_name as "name",
  p.fingerprint as "card_fingerprint",
  pgroup.transfer_ids as "transfer_ids"
FROM 
  (SELECT
    pd.payment_id,
    pd.fund_id,
    sum(CASE WHEN pd.amount > 0 THEN pd.amount ELSE 0 END) as "gross",
    sum(CASE WHEN pd.amount < 0 THEN pd.amount ELSE 0 END) as "refunded",
    sum(pd.fee) as "fee",
    array_agg(pd.payment_transfer_id) as "transfer_ids"
    FROM payment_details pd
    GROUP BY pd.payment_id, pd.fund_id) as pgroup
  JOIN payments p ON p.id = pgroup.payment_id
  JOIN funds f ON pgroup.fund_id = f.id
ORDER BY p.created_at
) to '/Users/troy/Projects/church/bayside_payments.csv' with csv header;