ASFA-Veda Data Benchmark

Transcription

1 ASFA-Veda Data Benchmark March 2015

2 The Association of Superannuation Funds of Australia Limited (ASFA) Level 6, 66 Clarence Street Sydney NSW 2000 PO Box 1485 Sydney NSW 2001 T F W (outside Sydney) ABN ACN This material is copyright. Apart from any fair dealing for the purpose of private study, research, criticism or review as permitted under the Copyright Act, no part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise without prior written permission. For more information To find out more about the ASFA-Veda Data Benchmark, including how your fund can participate, please contact Steve Shoreson from Veda. Steve Shoreson Head of Wealth Data Integrity, Veda T (03) E Steven.Shoreson@veda.com.au ASFA 2015 March 2015

3 1. Key points The 10th edition of the ASFA-Veda Data Benchmark analyses 3 million active members across 26 funds, 7 funds with less than 10,000 members, 11 funds with between 10,000 and,000 members, and 8 funds with more than,000 members. At 88.1 per cent, the ASFA-Veda Data Benchmark indicates that almost 12 per cent of member records have important core data that is missing or incorrect. When excluding errors in member beneficiaries, the Benchmark increases to 91.1 per cent. This measure allows the comparison of Australian funds to United Kingdom (UK) funds. Australian funds are not achieving the data-quality standards set by the UK Pensions Regulator (TPR). See the March 2013 ASFA-ITM Data Benchmark report for more information. Self-administered funds tend to have better data quality than third-party administered (TPA) funds. The data integrity of large funds (over,000 members) at 87.8 per cent is under that of small-medium funds at 89.7 per cent. Measurement leads to improvement. Eight out of nine funds in the current ASFA-Veda Data Benchmark, which have undertaken repeat audits, have improved their data quality. Trustees now have CPG 235 as a guide to help improve data quality for fund members. This should encourage greater participation in the ASFA- Veda Data Benchmark. The ASFA-Veda Data Benchmark results continue to indicate that funds need to do significantly more to improve the quality and integrity of their member data. However, more funds are benchmarking data quality, with 26 funds/products included in the past 12 months. ASFA-ITM Data Benchmark March of 8

4 2. March 2015 ASFA-Veda Data Benchmark The Benchmark was established in 2010 to track industry improvements and highlight the issue of data integrity across a number of areas. Through the Benchmark, ASFA and Veda have been taking a leadership role for a number of years in terms of data quality and building awareness of the need for improvement in this area. The results continue to indicate that funds need to do significantly more to improve the quality and integrity of their member data. Until this edition, the ASFA-Veda Data Benchmark has been based on data audits of funds over a rolling two-year period. This edition represents a move to only including participating funds over a 12-month period. This will enable the benchmark to better reflect improvement/deterioration in data quality over a more recent period. Core member details The core member details Benchmark encompasses the actual non-transactional data fields mentioned in SuperStream (full name, date of birth, current address and tax file number [TFN]) and other member-specific nontransactional fields (date joined fund, beneficiary and gender) that APRA has linked to the prudential standards. All member fields specified as required from employers by the Australian Taxation Office (ATO) are tested. Figure 1: Results of core member details Benchmark ASFA-Veda Data Benchmark (March 2015) Percentage of fund members with no errors 95-% A -94% B 85-89% C 80-84% D 75-79% E Industry average excluding beneficiary Very good data with minimal cleansing requirements Core member details Benchmark The latest Benchmark results reinforce the need to improve data quality and integrity for superannuation fund members % F Less than 70% G Very poor data with high cleansing requirements Results of specific member detail fields audited Of the key static data fields, most funds have issues with TFNs, addresses and member beneficiary data fields. The following graph provides a breakdown of the actual clean member data by each data field tested. Figure 2: Results of specific member detail fields audited (percentage of members with no errors) Benchmark % DOB Date TFN Address Gender Beneficiaries Name BENCHMARK Benchmark joined fund excluding beneficiary Date of birth, TFN, address and full name are the critical fields needed to confirm the correct identity of members for the correct application of contributions, communications, reporting and release of benefits. Full commentary on the importance of each of the member fields tested is provided in the September 2012 Benchmark Report, which is available from the ASFA website. ASFA-ITM Data Benchmark March of 8

5 The top ten funds (as at March 2015) The ASFA-Veda Data Benchmark is an anonymous Benchmark and funds are only mentioned if they give their permission. The top ten funds are a mixture of small, medium and large funds. Each of the top ten funds performed above the benchmark of 88.1 per cent. Figure 3: The top 10 funds (percentage of members with no errors) Benchmark % Small funds with less than 10,000 members Medium funds with 10,000 to,000 members Large funds with over,000 members ASFA-ITM Data Benchmark March of 8

6 3. Data cleanliness by ASFA-Veda Data Benchmark groups There are 26 funds now participating in the Benchmark. The average size of the 26 funds audited by Veda over the past 12 months (by number of active members) is 114,167. The following table provides the Benchmark for funds by category. The Benchmark excluding beneficiaries field provides a good comparison of the core data fields specified by the ATO with the data quality activity being undertaken by the UK Pensions Regulator (see the March 2013 report for more information on the UK data quality standard). Figure 4: Fund performance in terms of Benchmark fields Funds by benchmark group Small <10,000 members Medium <,000 members Large >,000 members TPA Selfadministered Number All funds Total membership 22, ,600 2,475,772 1,849,451 1,118,879 2,968,330 Average membership 3,280 42, , , , ,167 Date of birth 99.8% 99.8% % 99.8% 99.6% % Date joined fund.0% 99.9% 98.5% 98.3% 99.4% 98.7% Tax file number 97.3% 95.2% 95.1% 94.3% 96.6% 95.1% Address 97.6% 96.7% 97.5% 97.1% 97.7% 97.3% Gender 99.9% 99.9% 99.2% 99.0% 99.9% 99.4% Beneficiaries 84.0% 96.5% 96.6% 95.5% 98.1% 96.6% Name 99.9% 99.5% 99.8% % 99.8% % ASFA-Veda Data Benchmark Benchmark excluding beneficiaries 88.2% 89.8% 87.8% 85.8% 91.9% 88.1% 94.9% 92.5%.8% 89.5% 93.8% 91.1% Figure 5: Benchmark for specific groups of funds Benchmark % Small <10,000 members Medium <,000 members Large >,000 members TPA funds Self-administered All funds ASFA-Veda Data Benchmark Benchmark excluding beneficiaries Most self-administered funds have higher data quality than third-party administered (TPA) funds. Funds with less than,000 members tend to have cleaner data than larger funds. However, as always, there are some exceptions to this. ASFA-ITM Data Benchmark March of 8

7 Data integrity by fund size For the purpose of this report, small-medium funds (18 in the Benchmark) are defined as those with up to,000 members and large funds (8 in the Benchmark) are defined as those with more than,000 members. When comparing the small-medium and large funds as two groups, the Benchmark result is generally better for smallmedium funds. Figure 6: Small-medium funds versus large funds (percentage of members with no errors) DOB Date joined fund TFN Address Gender Beneficiary Name ASFA-Veda Benchmark Benchmark w/o beneficiary Small-medium Large All funds Self-administration versus TPA Self-administered funds have fewer members with data issues than TPA funds. The only area where the TPAs bettered self-administered funds was in data of birth (0.2 per cent better). The major differences above a 2.5 per cent variance between the two groups are that self-administered funds have better data quality for TFNs and member beneficiary record-keeping. Figure 7: Self-administration versus third-party administration (percentage of members with no errors) DOB Date joined fund TFN Address Gender Beneficiary Name ASFA-ITM Benchmark Benchmark w/o beneficiary Self-administered TPA All funds ASFA-ITM Data Benchmark March of 8

8 4. Benchmarking produces improved data quality APRA has provided trustees with a framework to help them mitigate data risks and improve data quality. This is represented in CPG 235 Managing Data Risk and supervisory visits from APRA now have a focus on data integrity. Funds participate in the Benchmark for a number of reasons. APRA s push for improved data quality is one. Many funds first participate as part of data integrity checking exercise when significant events are occurring (such as a change of administrator or a change of platform). Once they audit their data, the benefit in repeating the process becomes obvious. Figure 8: Improvement with repeat data audits (percentage of members with no errors) Benchmark % Fund A Fund B Fund C Fund D Fund E Fund F Fund G Fund H Fund I Run Latest run Benefits of repeat data audits There are nine funds in the current ASFA-Veda Data Benchmark that have performed more than one data integrity audit. Eight out of nine have improved overall data quality. The one exception showed a small deterioration of 1.2 per cent in data quality. Of the eight funds showing an improvement in the quality of member static data: the smallest improvement in data quality was 1.7 per cent and the two largest were 22 per cent and almost 50 per cent a further four funds improved their data quality by between 10 per cent and 20 per cent. This signifies that data cleansing is taking place where errors are reported. Value for funds and members Every fund that has participated in the Benchmark more than once has seen a continuous improvement in data quality, leading to a better member experience and: a reduction in missing and/or incorrect data less complaints from members (and a reduction in errors and omissions) better communication through improving member contact details. APRA recommends that trustees benchmark their own improvement and also against other superannuation funds. A clear conclusion that can be drawn is that once trustees begins to participate in data auditing and benchmarking, the majority of funds will realise a significant improvement in data quality for their members. Good quality data is more than just a regulatory or governance concern it is the bedrock for being able to implement effective data driven customer engagement and digital strategies. ASFA-ITM Data Benchmark March of 8