Supporting information:
Historical data (Excel spreadsheet)Survey notes for table C15
Chronology of changes to tables and current status
December 2010: Review of sample
The Bank has reviewed the composition of its quarterly and annual managed funds surveys. In order to reflect changes within the managed funds industry over the last few years, four fund managers have been brought into the quarterly survey and data has been backdated to December 2003 (where appropriate). This has increased reported quarterly funds under management slightly over this period.
September 2010: Series break
Owing to a change in the nature of its ownership, a respondent, previously defined as a fund manager for the purposes of the Managed Funds Survey, has been excluded from the survey from the September quarter. This has resulted in a break in the managed funds statistics of around $400 million for total funds managed in the September quarter.
June 2010: Revisions to total funds under management (FUM) and presentation changes
Table C15 is now presented horizontally broken down first by asset domicile and then by asset classes. New data provide a breakdown of overseas assets into holdings of securities, equities, or other assets from June 2010. Specific information on household FUM and non-household FUM is also provided.
A proportion of FUM previously reported as life insurance has been transferred to ‘other funds managed’, in a revision backdated to December 2003. This change leaves the life insurance aggregate representing assets backing ‘policy holders’ funds’. The proportion transferred constitutes reserves not yet attributed to policyholders, and other assets. This change has increased the total of non-household FUM and reduced household FUM by a corresponding amount.
March 2008: Revisions to total FUM and presentation changes
Total FUM have been revised down on average $400m for each quarter since December 2003. Superannuation FUM have increased on average each quarter by around $1 billion, with a corresponding fall in unit trust and other funds managed. Total equity and property FUM in New Zealand are the primary source of the net total FUM reduction, an increase in cash and deposits has been offset by a corresponding reduction in short-term securities.
Life insurance and superannuation FUM now appear in one column only. The former presentation is retained in an historical table showing December 2003 to December 2007 (this presentation is not updated with this or subsequent releases). Alternative presentations for the primary C15 table (tables C15a, C15b and C15c) have been introduced to assist analysis.
December 2007: Amendment of survey scope and introduction of KiwiSaver data
The number of funds in the survey has increased. Three funds were removed from the survey and four, previously surveyed annually, were added. Quarterly imputation for these funds means there are no series breaks for this reason over the period. A minimum of $500m in FUM at December 2007 is the criterion for inclusion in the quarterly survey. FUM in the quarterly series account for approximately 90 percent of all FUM assessed at December 2007, when a supplementary annual unpublished survey is undertaken. The full survey results for all household FUM are disseminated on the Bank’s website in May each year. (Around $5 billion of total FUM were managed on behalf of charities and other non-household sources in 2007.)
FUM are defined by net asset value and include only those assets managed directly by the fund on behalf of retail and wholesale clients of the fund (e.g. excluding listed vehicle values other than any in which the MFS respondent’s clients’ funds are invested). Double-counting among funds is excluded.
Revisions to existing data have also been made. As a result, for each quarter since December 2003, the level of household FUM is approximately 1 percent higher. This increase is in part due to re-allocation of FUM from non-household to household origin. A moderate reduction in the proportion of FUM invested in property has occurred, and the allocation of property assets has fallen in the unit trust category and increased slightly in other product categories.
KiwiSaver FUM with quarterly survey respondents and the Inland Revenue Department are included in superannuation data from and including December 2007
June 2007: Revisions to other funds managed
Total funds under management for June 2007 include approximately $500m in the 'other funds' category not previously captured in the survey
December 2003: Survey coverage increased and series break
The survey population for the quarterly survey increased from 17 to 28 reporting entities for December 2003 (some ‘fund managers’ comprise more than one reporting entity). The additional fund managers introduced to the survey this quarter are listed in the background notes to the series.
The Bank surveys FUM comprehensively at December each year, using an unpublished annual survey to supplement information from the quarterly series. The comprehensive survey at December 2002 showed that the quarterly series was measuring approximately 80 percent of FUM. Greater quarterly coverage is required, and 11 funds from more than 40 in the annual survey have been brought into the quarterly survey. This brings it close to an estimated 90 percent of values obtained from the full December ‘census’ coverage.
Series break
There is a break in the quarterly series between the September and December 2003 quarters. The following information is provided to assist with analysis of December 2003 data.
Total FUM for the 17 ‘core’ funds grew in the year to December 2003 by five per cent. Total FUM for these funds increased more than $1 billion from the September to the December quarter 2003.
The 11 funds introduced to the quarterly survey in December have had annual growth in FUM to December 2003 of 18 percent. They have brought $1 billion of new funds to the unit trust category, over $0.5 billion to employer superannuation and over $4.5 billion to ‘other funds’. The newly-introduced funds tend not to manage client funds using unit trust, superannuation or life insurance products, which accounts for their disproportionate representation in the ‘other funds’ category.
For the 28 funds reporting, FUM grew $3 billion in the year to December 2003, or seven percent.
FUM include non-household financial assets
Not all funds under management are sourced from the household sector. In December 2002 over $4 billion of the FUM obtained from the quarterly and annual surveys combined were from charitable, non-profit and commercial sources. The December 2003 quarterly data alone includes over $4 billion from these sources.
The long-run household assets and liabilities series, which includes data as at December for managed fund assets of households, does not provide data on other sectors providing FUM. There is no long-run estimate provided for non-household FUM, but in recent years it has represented around 10 percent of total FUM.
December 1999: Revisions and improved classification of assets
The December 1999 release included widespread data revision to earlier quarters, from March 1999 onwards. Two main changes occurred. First there was a change to product columns in table C14 [discontinued in December 2002], and associated assets in table C15. The principal shift was from ‘unit trust’ and ‘other’ classifications to ‘retail superannuation’, with about $600m being added to this category. Second, a lesser amount (from about $140m to $200m over the year) was added to unit trust totals, in the overseas asset category. There were also in aggregate many minor adjustments to asset classifications, incorporating improved information from funds surveyed that became available over the year. Note that these revisions are intended as far as possible to present the four 1999 quarters on a ‘like for like’ basis, so that intra-quarter changes represent changes to business results for the fund managers listed at the end of this note.
From 1999, table C15 classifies funds managed by asset category in New Zealand, as before, but with improved definitions of these categories, which has resulted in markedly changed results in the case of ‘cash and deposits’ and ‘short-term securities’. Also, the former survey obtained a breakdown of asset categories for overseas investments, while the total alone is now obtained by the Bank. Statistics New Zealand collects information on asset categories.
March 1999: Survey coverage increased and series break
Change to sample size
The data for the Managed Funds Survey (MFS) from March 1999 onwards are drawn from a subset of the managed funds that contributed to the quarterly MFS survey that the Bank first began publishing regularly from March 1996. The revised survey represents an evolution in the collection of managed funds data. This principally reflects the fact that increasing concentration in the industry has enabled the Bank to obtain well over 90 percent of 1996-1998 totals from fewer than 20 fund managers, compared to the 60-odd surveyed until September 1998. From March 1999, two relatively large fund managers that were not previously in the survey were added, with the result that the total of funds under management reported from March 1999 increased slightly.
The changes that occurred at the end of 1998 in the sample population of the survey, and revisions to data requirements precluded backdating. The survey data regularly published from March 1996 to September 1998 were based on an initial data set agreed with the managed funds industry in 1995. A first report of the results of the survey, for the September quarter 1995, was published in November in a Statistical Discussion Paper, S95/1. The regular quarterly series then began from March 1996. It was noted in S95/1 that, as with other new series the Bank had introduced, both the industry and the Bank would work together to improve the quality and relevance of data collected over time. The March 1999 revision represented a step forward in the process of increasing the value of information available about the managed funds industry.
Classification changes
In addition to changing the sample size for the survey, the Bank clarified data definitions in the light of experience and international standards. In particular, regard has been paid to Australian practice, as many fund managers are subsidiaries of Australian, or global, operators with significant FUM in Australia. The key organising principle for the survey has been to obtain FUM classified by the tax regime applied to fund products as they are presented to the fund customer. The tax regime for unit trust, life company, and superannuation products is different, and the organising principle for the data collection is to track fund growth in this context. FUM are classified as superannuation funds if they belong to a registered superannuation scheme. Group Investment Funds (GIFs) have two classes with different tax treatments, A and B under the Trustee Companies Act 1967, but confidentiality concerns again preclude publishing the distinction, and they are included with unit trust data. The ‘other funds’ column presents funds under management that have not been obtained through the other specified tax regimes.
Former column headings for products sold with a life insurance element were ‘Investment-linked’ and ‘Non-linked’ life. They were changed to ‘unitised products’ and ‘non-unitised products’ to try to improve intuitive comprehension of the underlying meaning. Unitised products include all life bond investments. This category is also intended to include a class of product, offered with a life insurance element, which provides a return that is not adjusted through the same kind of actuarial process that allocates bonuses and other annual returns to traditional whole of life and endowment policies. The latter are ‘non-unitised’.
Superannuation funds under management in the survey until September 1998 were requested under headings ‘wholesale’ and ‘retail’. Readers and many survey contributors assumed this meant ‘employer’ and ‘non-employer-related’ superannuation funds. Some contributors however reported funds received for management from a retail (non-employer-related) source as ‘wholesale’, on the basis of the commercial relationship. The distinction now obtained is whether funds being managed are originally from a registered employer scheme (irrespective of whether there is an employer contribution), or otherwise, in a registered (retail) superannuation scheme.
December 1998: No data available
It has not proved possible to publish data for December 1998.
September 1998: Reclassifications
The main change from the survey results to September 1998 is that ‘cash and deposits’ is clearly defined as ‘deposits’, to exclude short-term securities such as registered certificates of deposit and other short-term commercial paper. Cash and deposits are now around a third of the previous total, with short-term non-government interest-bearing securities correspondingly increased. Note however that some funds invest in ‘synthetic’ 90-day bank deposit products that are not tradeable short-term securities, which to some extent accounts for the relatively high ‘cash and deposit’ ratio that continues to be reported. A related change includes a clear specification that interest-bearing securities are reported by time to run to maturity (residual maturity), not original term of issue. The break between short- and long-term occurs at one year.
