Privacy Law and Policy Reporter
The release of the fourth set of reports on the Commonwealth's data-matching programs confirms earlier conclusions that revenue from the programs has fallen far short of original expectations. The Data-matching Program (Assistance and Tax) Act 1990, which initiated the Parallel Data-matching Program (PDMP), requires annual reports on the program from each of the five agencies involved in the program. The reports for 1994 confirm trends identified in Clarke's earlier review of the PDMP ((1994) 1 PLPR 8), which led to his conclusion that the program has proven cumbersome and ineffectual. It should be noted that Clarke's review casts doubt on the accuracy of the processes used to calculate costs and benefits ((1994) 1 PLPR 9).
Data-matching is conducted in cycles and includes the matching of identity, income and payments. During 1993-1994, five matching cycles were conducted on 72.4 million records, leading to 2.7 million discrepancies (3.8 per cent). Yet only a fraction of these led to completed reviews; the DSS, for example, completed reviews on only 22,622 of the 1.94 million discrepancies identified through data-matching, identified net gains for the departments were $63.7 million for Social Security, $4.2 million for DEET (down from $4.3 million in 1992-1993), $4.1 million for DHRD, $2.8 million for DVA, and a $3.2 million loss for the Tax Office.
The report also continues the well-established trend of downward revisions to expected net benefits. The benefits from data-matching peaked in 1993-1994 and are expected to decline over the years ahead. DSS projects the net gains of the PDMP to fall to $39.8 million in 1994-1995, $23.9 million in 1995-1996, and to $21.2m in 1996-1997 and 1997-1998 - by which time the ratio of costs to direct benefits will have fallen to one to two. This is well short of the $300m annual gains projected at the introduction of the program in 1990. These figures are lifted only by estimating an annual $90m benefit from 'voluntary compliance'.
Included in the Tax Office document is a report of a recent data-matching exercise which reflects problems of the data-matching process. An assessment of pensioners receiving both a foreign and Australian pension produced 86,670 discrepancies between social security and declared taxation information. Further sampling showed that the problem was not fraud or dishonesty, but that the 97 per cent of these pensioners declared this income in other places on their taxation returns or did not exceed the tax threshold. The exercise did not lead to any identification of undeclared income or recovery of payments.
Other changes this year include the Tax Office's capacity to specially monitor 'compromised' Tax File Numbers (TFN) where there is evidence to suggest that the TFN may have been disclosed to an unauthorised third party. The Tax Office also launched its controversial Schools Project, which besides giving general tax information to students aims to have 70 per cent of year 10 secondary students applying for TFNs in the school environment. Although these students may not be paying tax for some years the PDMP makes it necessary for them to have a TFN for educational and job seeking assistance and for the contributions system in post-secondary education.
The reports also note the continuing involvement of the Privacy Commissioner's office in auditing data-matching programs in line with the Commissioner's guidelines. One result of this during 1993-1994 was that the DSS launched a major effort to remove tax file numbers from paper files on individuals in regional offices.