How to display 1 column in 2 fields on the report

If 3 columns are static (meaning that no account has more than 2 email addresses then you can use the static PIVOT but in order to use it you first need to come up with the common data for all accounts. Row numbering partitioned by the ID would be a good choice. Here is the sample:

select
    pvt.ID, pvt.[1] as [Email I], pvt.[2] as [Email II]
    from 
    (
        select 
            ID, Email, row_number() over (partition by ID order by ID) N
            from TblB
    ) src
    pivot (max(Email) for N  in ([1], [2])) pvt;

This will work if you have SQL Server 2005 or better.

<!-- Begin Edit

For restating the query in Oracle dialect, the version is very important. As @Pavel Pawlowski mentioned in his comment, CASE is available in Oracle 9i or better, so it can be used in place of a more traditional DECODE. Below are the scripts for both newer and older versions of Oracle. I restated both as left join because I believe that it makes sense in a scenario when some records in the parent table named TBLB do not have the match in the child table TBLA, but the IDs from TBLB still need to be included in the results. Another important detail to point out that the names of the database objects should not contain spaces, the names are not case sensitive, but the column values are always case sensitive. This means that the words 'work' and 'home' should be either spelled in correct case or else the LOWER function should be applied to the column. Here are the scripts (one for each version):

/* This script should work in newer versions of Oracle, 9i or better */
SELECT
    TBLB.ID, MAX(
    CASE 
        WHEN mail.EMAIL_CODE = 'work' then mail.EMAIL_ADDRESS 
        ELSE NULL 
    END) WORK_EMAIL_ADDRESS, MAX(
    CASE 
        WHEN mail.EMAIL_CODE = 'home' then mail.EMAIL_ADDRESS 
        ELSE NULL 
    END) HOME_EMAIL_ADDRESS
    FROM TBLB LEFT OUTER JOIN TBLA AS mail
        ON TBLB.ID = mail.ID AND mail.EMAIL_CODE IN ('work', 'home')
    GROUP BY TBLB.ID;

/* This script should work in older versions of Oracle, 8i or worse */
SELECT
    TblB.ID,
    MAX(DECODE(mail.EMAIL_CODE, 'work', 
        mail.EMAIL_ADDRESS, NULL)) WORK_EMAIL_ADDRESS,
    MAX(DECODE(mail.EMAIL_CODE, 'home', 
        mail.EMAIL_ADDRESS, NULL)) HOME_EMAIL_ADDRESS
    FROM TBLB,
    ( 
        SELECT 
            ID, EMAIL_CODE, EMAIL_ADDRESS
            FROM TBLA WHERE EMAIL_CODE IN ('work', 'home')
    ) mail
    WHERE TBLB.ID = mail.ID(+)
    GROUP BY TBLB.ID; 

So the only problem with the original script in question was that the GROUP BY was included more columns than it deserved and it was also in the wrong place.

End Edit -->

Hope this helps,

Oleg