September 1997

Insurance

and

Superannuation

Bulletin

WTO agreement on financial servicesWorld Trade Organisation (WTO) negotiations on Financial Services which were concluded in Genevaon 12 December 1997 has resulted in a landmark agreement covering more than 95 per cent of world

trade in banking, insurance, securities and financial information. In all, 102 members of the WTO nowhave multilateral commitments in this sector.

The WTO is an inter-governmentalorganisation created in 1995 as the successorto the General Agreement on Tariffs and Trade(GATT) organisation. It is the onlyinternational agency overseeing the rules ofinternational trade. Since its creation, theWTO has been the forum for successfulnegotiations to open global markets intelecommunications and in informationtechnology equipment. Financial Services cannow be added to that list.

Announcing the new agreement the DeputyPrime Minister and Minister for Trade, MrFischer, said the new trade agreement onglobal financial services was 'a victory forAustralian businesses and a victory forAustralian trade policy'.

'This WTO agreement will mean the Australianfinancial sector will be able to invest moreeasily, with more confidence and more securityin the Asia-Pacific region and across theworld,' Mr Fischer noted.

The signatories to this historical agreement willhave until January 1999 to ratify theircommitments and must implement themeasures by March of that same year. Theoutcome reached in Geneva is also importantin terms of achieving future liberalisation,particularly in the WTO negotiations onservices commencing in the year 2000.

Particularly pleasing from an Australianperspective was the positive attitude towardsliberalisation adopted by our neighbourcountries in the Asia-Pacific region. At a timeof considerable financial, and in some casespolitical, turmoil these countries werenonetheless prepared to make significant offersto improve access to their financial sectors.This reflects both the recognised importance offree trade flows worldwide and theunderstanding that the only long term solutionto re-establishing stability in the region isthrough further liberalisation of financialmarkets.

Two brief examples serve to demonstrate theapproach taken by the Asia-Pacific countries:

• Malaysia is to allow foreign ownedinsurance companies to hold aggregateshareholdings of up to 51 per cent in

domestic companies - an offer which wouldguarantee continued investment byAustralian companies including QBE. Newprovisions allowing foreign participation ina range of financial activities and thegranting of additional re-insurance licencesalso formed part of the Malaysiancommitment.

• Similarly, in Indonesia (where both AMPand ANZ are major investors) overseascompanies now have guarantees thatrespect current conditions of ownership andpermission to take up to 100 per cent equityin publicly listed insurance and other non-bank financial institutions.

Australia has played a key role over manyyears in securing the successful outcome ofthese negotiations. It is a generally held viewthat the Prime Minister's announcement ofAustralia's strengthened WTO offer at theAPEC conference in Vancouver providedimpetus for other nations to seek a positiveoutcome in Geneva.

The Australian offer reflected the very opennature of the Australian market for financialservices but was conditional on our key tradingpartners, both developed and developing,undertaking significant liberalisation in thenegotiations. This set the scene for other WTOmembers to think seriously about their offers.

The Government believes internationalcompetition acts as a catalyst to lowering thecost and increasing the range of financialproducts and services available to thoseseeking to do business in Australia withoutsacrificing the integrity or stability of theAustralian financial system.

The WTO agreement complements theGovernment's recent decisions, flowing fromthe Wallis Inquiry, regarding the reform of thefinancial sector. Australia will have a morestable, more competitive and more efficientfinancial system that will not only bepositioned to compete strongly in the globaleconomy, but will result in lower costs and anincreased range of products and services onoffer for those seeking to do business inAustralia or to use Australia as a focal point forregional activities.

These initiatives will ensure Australia'sfinancial system provides the best possiblefoundation for the continued development andgrowth of the economy and will put Australianinstitutions in a sound position to takemaximum advantage of the improvedinternational market opportunities flowingfrom the WTO agreement.

ISC supports Australian Investment PerformanceMeasurement and Presentation Standards (AIPMPS)

The AIMPS have now been published in their final form ready for the commencement date of 1 January1998. These important standards establish a common methodology for measuring and presentinginvestment performance in the wholesale marketplace and have been developed by an industry workingparty after extensive consultation and review.

The ISC has maintained a close interest in the development of the standards throughout and congratulatesall involved in attaining this landmark. As well as the significant work put in by the industry, the ISC isalso supportive of the self regulatory approach which has been adopted. While the standards themselves arefairly technical, the progress so far illustrates how a self regulatory approach can be developed even insuch a complex area.

The existence of the standards should greatly assist users of performance information, particularly trusteesof superannuation funds. A common basis for the preparation and presentation of results means that theusefullness of the information can be maximised and made more effective.

Whilst not mandatory, the existence of the standards means that trustees will be able to ascertain whetherthey are being provided with information in accordance with the standards and, if not, can be aware of therisk that they will be less able to compare investment managers who are not reporting using the standardsthan those that do.

The ISC looks forward to following the progress of the standards and their implementation.

The four pillars supporting the insurance sectorInternational Monetary Fund (IMF) in a 1996 publication ‘Bank Soundness and Macroeconomic

Policy’, outlined four principles that help determine safety and stability in the financial system. Whilethe IMF authors, Carl-Johan Lindgren, Gillian Garcia and Matthew I Saal had banking in mind, the

IAIS considers that these principles have a wider application to the insurance sector.

Each country is concerned to have (amongother things) a safe and stable insurance sector,to protect policyholder interests in particularand to avert financial instability moregenerally. At the same time, governments arekeen to ensure that arrangements put in placeto provide safety and stability in the financialsector also promote competition, efficiency andinnovation.

An effective framework for a sound andefficient insurance sector has four pillars:internal governance, market discipline,prudential regulation and internationalcoordination. These four pillars should beregarded as complements rather thansubstitutes, as mutually reinforcing rather thanmutually inconsistent.

1. Internal governance

An insurance company practises good internalgovernance when its conduct is prudent, open,honest and fair. Companies should haveinternal structures and processes in place todemonstrate that the commitment to corporategovernance, and the certification ofcompliance, are occurring at the highest levels.

Internal governance is essential to publicconfidence in the insurance sector, because thesoundness of an insurance company is first andforemost the responsibility of its owners,directors and senior managers. Insurancecompanies should have in place a system ofinternal standards and controls to manage riskand encourage prudence in the conduct of thebusiness. There should be mechanisms forchecking and certifying compliance with thestandards and controls.

When an insurance company gets into trouble,and possibly fails, the cause can generally betraced to lax management, or in other words,poor internal governance.

Insurance is a long-term business, and aninsurance company’s owners, directors andsenior managers are responsible forcapitalising the business with sufficientresources to meet commitments, absorb shocksand remain viable. Each company needs toensure that its directors and senior managersare competent and ethical, are motivated to runthe business efficiently and prudently, and are

encouraged to keep the company in aprofitable, liquid and solvent condition.

Directors and senior managers should be madeaware that failure to practise good corporategovernance will expose them to the risk oflosing control of the business throughbankruptcy, takeover or official intervention.

Lax management in an insurance company cantake the form of incompetence, negligence orfraud. Internal pitfalls are less likely ifcompanies have effective policies andprocedures for monitoring and controllingrisks, and checking and certifying compliance.Internal controls would include measures suchas: ‘fit and proper’ tests for directors andsenior managers; techniques for measuring andlimiting exposures; and arrangements forinternal and external auditing.

Good internal governance is more likely tooccur when companies are also subject tostrong market discipline.

2. Market discipline

Market discipline can provide an incentive fordirectors and senior managers to not merelypay lip service to good corporate governance,but to actively practise and promote it.

In a competitive insurance market, consumers,creditors and analysts can reinforce theincentives companies have to operate safelyand soundly by monitoring performance,exerting discipline and, as a final step, forcingpoorly managed or unsound insurers out of themarket.

However, market discipline cannot work as aneffective means of keeping companies prudentin their business conduct if insufficientinformation is available to market participants,or if there is a widespread perception that theGovernment will always bail out an insurerwhich runs into serious trouble.

Disclosure of timely, comprehensive and clearinformation about an insurance company’sproducts, prices, profits and financialsoundness is essential if market discipline is tobe effective. While individual consumers maynot be well equipped to monitor companies onthe basis of published financial statements,creditors, analysts and competitors will be able

and keen to use the information to ratecompanies, and these assessments will in turnfeed into the company’s reputation andstanding in the marketplace.

Market discipline can also fail if investorcompensation schemes are excessivelygenerous, since companies are less risk averse,and customers and creditors are less vigilant,where there is a perception that troubledcompanies will be rescued by the authorities orlosses will be recouped from the government.Failing companies should be allowed to exit inan orderly manner without any guarantee thatpolicyholder losses will be fully compensated;otherwise there is a perverse incentive for poorinternal governance. Shareholders should, ofcourse, bear the losses.

While good internal governance and strongmarket discipline go a long way towardsencouraging efficiency and soundness, mostcountries consider that official regulatory andsupervisory oversight of the insurance sector isalso necessary to protect policyholders andmaintain public confidence.

3. Prudential regulation

Prudential regulation of insurance by theofficial supervisor is not a panacea; but it canbe an effective means of limiting the riskexposures of companies, and encouragingproper and prudent management of those risks.However, prudential regulation should bedesigned so as to avoid being anti-competitiveor commercially intrusive.

For example, entry barriers created in theinsurance sector for prudential purposes shouldnot be so high as to unduly limit competitionand protect large companies. Restrictingcompetition in this way rarely improvessoundness, since lax and lazy managementsthrive in protected environments.

Prudential regulation can provide assurancethat directors and senior managers arecompetent and ethical, that adequate riskmanagement systems are in place, and thatcompliance with the standards is monitoredand certified at the highest levels. Further tothis, prudential regulation results in thepreparation and publication of additionalinformation which can be used by the market(as well as the regulator) to discipline under-performing companies.

The role of prudential regulation in supportinginternal governance is to require and check thatcompanies institute adequate internal controlpolicies and procedures, and that senior

managers are familiar with and responsible forthe risk assessment and management process.While the fine detail of internal control can beleft to insurers’ own discretion, it is essentialthat accounting and auditing standards areadequate in enabling senior managers to detectand remedy weaknesses, and that regulatorsconduct on-site inspections to satisfythemselves about the quality of the company’smanagement and systems.

Finally, globlization and technology have inrecent years created new dangers for financialconsumers and new threats to financialstability. Responding to this requires anunprecedented degree of internationalcooperation and coordination among financialregulators, including insurance regulators.

4. International Coordination

Insurance groups are increasingly operating inmultiple jurisdictions. This raises the spectreof regulatory arbitrage by companies seekingbottom level standards, and of the potentiallyrapid transmission of shocks in crisissituations.

There are no international laws governinginsurance, and country based regulationinevitably results in inter-country differences inaccounting standards, in exit procedures, andin supervisory rules generally. Therefore,harmonisation of standards is desirable ifinternational financial stability is to beprotected from the threats posed by weakjurisdictions with lax supervision.

Reinforcement of market discipline at theinternational level has come chiefly in the areaof information disclosure. Improved disclosurestandards and additional information on thecondition of individual insurance companieswould better permit market participants toassess their financial soundness, paving theway for improved market discipline.

It is notable that there are no internationalagreements on closure standards for failingfinancial institutions. As a result, theresolution of a failed insurance group operatingin several jurisdictions could be a fragmentedand contentious process. Thus, exit policy asan adjunct to market discipline functions onlyat the national level, with internationalcooperation essentially occurring only on an adhoc basis.

While international standards are generallydesirable, there are some practical limitationsto full regulatory harmonisation. In particular,countries with high economic volatility and

permissive accounting conventions may bebetter served by stricter regulatory standardsthan are prescribed by international minimums.Also, arrangements for investor compensation(the official safety net) will vary from countryto country.

International coordination is not yetsufficiently developed to offset thesedifferences in economic conditions andregulatory environments across jurisdictions.Nonetheless, there is much value to be gainedin the identification and dissemination of bestpractices in insurance regulation, and evenmore so in the practical application of suchinternational standards to create a soundinsurance sector in each of the variousjurisdictions.

The International Association of InsuranceSupervisors (IAIS) is in the process ofdeveloping prudential standards for insurancesupervision. The standards would take theform of principles or guidelines of best practicethat insurance supervisors and regulators couldchoose to adopt and apply as they see fit intheir different jurisdictions. The standardswould not be black letter law: they would notbe compulsory, but there would be marketpressures for their application around theworld. The existence of IAIS standards wouldencourage regulators that have regulationsfalling short of the IAIS standards to improvethe quality of their regulations and supervision.Improved prudential regulation internationallywould help strengthen the global financialsystem.

Y2K - the millenium bug and superannuationThe year 2000 is rapidly approaching and the ISC is concerned many superannuation funds and

insurance companies may not be adequately prepared to manage the millenium computer bug. Thisarticle contains the main features of an ISC guidance note that was recently released to super funds

advising them of their responsibilities about this issue.

The main issue regarding the Y2K problem - themillenium bug - is to ensure that your fund’s orcompany’s computers can accurately record,manipulate and interpret data and produceaccurate outputs from the Year 2000 onwards.Many computer systems, especially older onesand new systems dependent on older or legacysystems, will need their program logic to bemodified. Normal business operations could besignificantly disrupted. An additionalcomplication is to provide for 2000 being a leapyear.

Trustees and funds which rely on vendors formany applications may need to replace theirhardware and operating systems in addition toupgrading their software. This is becausesoftware upgrades which are Year 2000compliant may not run on old hardware, or mayrun but with significantly adverse implicationsfor achieving agreed service standards.

The ISC’s primary concern is protecting theinterests of members from suffering anydetriment due to any deficiencies in resolvingYear 2000 issues. As such, the ISC considers itis critically important to have all systems whichwill affect fund members to be ready and testedwell before the Year 2000 (ideally by the end of1998). Examples of these include accountingsystems, administration systems, investmentsystems and systems for communicating withmembers.

Dependencies on, or links with, external partiesare also important when considering Year 2000issues. These external parties include employersponsors who may communicate electronicallywith the trustee or administrator, custodians,administrators, investment managers, mailinghouses and employer sponsors. Should any ofthese external parties have significant problemsdealing with Year 2000 issues it could haveflow on effects to the superannuation fund.Where funds rely upon information conveyedelectronically from such sources they shouldinsist on written assurances being provided bythe party responsible. Trustees should expectexternal service providers to have undertaken a‘computer audit’ to assure themselves that theirsystems are Year 2000 compliant andcompatible with those of their customers. Theissue of compatibility of solutions is critical toensure that the approach adopted to resolve

identified issues will ultimately work in respectof the fund.

The Year 2000 issue will affect more than justmajor computer systems. Problems may alsoarise with applications such as spreadsheetswritten and maintained by individuals within thetrustee or administrator. In addition, non-ITsystems such as security systems andcommunication networks (utilising embeddedchips) could have trouble coping with the Year2000. Notwithstanding appropriate forwardplanning and strategies to resolve this issue it isalso vital that disaster recovery programs areupdated to deal with any contingencies whichmight arise.

The scarcity of technical resources is animportant issue when estimating the timing ofpreparations for and the cost of the Year 2000problem. As the Year 2000 draws closer thiswill become an increasing problem with thepotential to cause costs to increase significantly.

It is also extremely important to allow adequatetime for testing and re-testing programs. Everysystem component which is made compliant hasto be tested alone and in interaction with allother system components. Tests of interactionwith external parties have to be coordinated.Due to the need for ongoing operations of thefund and external parties, the time available fortesting may be restricted to weekends.Interrelated applications have to be coordinatedas to when they go into production.

Finally, while appropriate legal safeguardsshould be incorporated in agreements withexternal service providers, anticipation andprevention of problems rather than settlement ofdisputes, should be the focus of trustees’attention.

Action plan

In its paper titled “The Year 2000 - A Challengefor Financial Institutions and Bank Supervisors”the Basle Committee on Banking Supervisionoutlines the steps institutions need to follow toresolve the Year 2000 problem.

Although the focus of the Basle paper is on thebanking industry, the ISC believes the mainrecommendations of the paper can be applied tothe superannuation industry through thefollowing six steps:

1. Developing a strategic approach

This phase includes establishing Year 2000 as astrategic objective at the highest level within theorganisation, developing a process tocommunicate the strategic objective throughoutthe organisation, and assessing the resourceimplications of the Year 2000 at a very highlevel.

Organisations should now be well past thisphase in addressing issues. This phase shouldhave entailed the development of a properbusiness plan for this problem - some possiblestepping stones which could be used in such aplan are:

• Prepare a strategic plan to underpin Year2000 activities on a 'whole of group' basisfor the critical period ahead. The planshould include a full risk analysis, keyobjectives and milestones, performancemeasures and desired outcomes, and acquitalof performance against the plan in the annualreport/ accounts

• Develop a model to reliably and accuratelyestimate Year 2000 costs

• Examine guidelines which already exist andwhich can be used to assist with themanagement of purchase risks arising fromcommon use and unique supplierarrangements for IT and non-IT goods andservices

• Seek written assurances about Year 2000compliance from suppliers

• Document the Year 2000 assessment andplanning activity to allow an assessment ofthe appropriateness and sufficiency of theapproach from a whole-of-businessperspective

• Establish appropriate mechanisms to ensuregovernance of the Year 2000 activities andto provide appropriate assurances tostakeholders in relation to the businessimplications of the Year 2000 problem.This should include assurances about the useof resources, efficient operations, financialintegrity and validity, compliance withlegislation, accountability to members andother stakeholders

• As part of an overall risk managementapproach, develop and document a strategyfor the management of Year 2000 risks thatincorporates an analysis of the operating,compliance and external environment whichare potentially affected by the Year 2000problem, the identification of possible

sources of risk, an assessment and ranking ofrisks, options for the treatment of identifiedrisks, and measures to monitor and reviewidentified risks.

• Review the Year 2000 project managementand take remedial action to ensure theproject is governed by clear and achievablemilestones which are regularly reviewed, allstaff responsible for aspects of the Year2000 project operate within clear authoritiesand lines of accountability, and the projecthas top level commitment, including foradequate resources and strategic direction.

2. Creating organisational awareness

Making certain that the strategic importance ofthe Year 2000 project as a business objective isunderstood and appreciated throughout theorganisation may be the most important phase inthe action plan. The recognition that this maybe a survival issue, requires not only a visiblecommitment from top management for itssuccessful resolution as a strategic priority, butalso an awareness of its importance by staff atall levels. Line management needs tounderstand the issue and its implications andaccept ownership of the issue. Responsibilitiesshould be clearly assigned. This phase has fourobjectives: creating visibility; ensuringcommitment; identifying resources; andspecifying specific strategic objectives at abusiness line level. Organisations should alsobe past this phase now in their Year 2000project.

3. Assessing and developing detailed plans

This phase moves from project concept toconcrete actions. Detailed inventories of whatmust be done are developed, coveringcentralised and decentralised hardware,software, and networks as well as equipmentembedded with computer chips and logic. Theinventories should include all aspects ofbusiness line activities whether internal to theorganisation or external to it. Risks should bequantified and priorities set based on these risks.

Organisations are expected to have completedthis phase or be very close to completing it bynow otherwise they are unlikely to meet theultimate deadlines.

4. Systems, applications and equipment

During this phase, the necessary fixing ofoperating systems, applications, hardware andequipment takes place. The development ofcontingency plans that identify alternativeapproaches if renovations lag or fail is animportant part of this phase.

Organisations should be well into this phase atthis time. Renovation work for significantapplications that need to be tested with thirdparties must be completed with enough time toallow for thorough testing. Completion of thishigh priority work should be targeted for mid-1998. Typically all renovation work would betargeted to be completed no later than the end of1998.

5. Validating the renovations

Testing represents the largest single task in theYear 2000 project. Detailed test schedules mustbe developed and coordinated with third parties.Data flows, internally and with third partiesmust be thoroughly tested while both the senderand receiver simulate Year 2000 conditions.

At least for larger institutions and all significantapplications, the validation phase should betargeted for completion by the end of 1998. Allvalidation work should be completed by mid-1999. Only with this schedule will there besufficient time for industry wide testing with allthird parties during 1999.

6. Implementing tested, compliant systems.

Implementation requires careful planning tomake sure that interrelated applications arecoordinated as to when they go into production.This implementation phase also requiresmonitoring of progress by service providers andvendors.

Industry questionnaire

In July 1997 the ISC sent a questionnaire onYear 2000 computer issues to all administratorswho look after 100 or more funds (about 200administrators in number) plus all ApprovedTrustees (about 180 in number). From themailout the Commission received 142responses.

The responses were split into two groups. Thefirst group are the smaller agencies comprisingthe small to medium superannuation funds,accountancy practices, administrators and creditunions. The second group comprises largersuperannuation funds, banks and keyadministrators.

Smaller organisations

There were 67 responses placed in this category.Despite the questionnaire specifically asking forfurther details by way of attachment theseresponses were typically just the return of thequestionnaire without detailed covering lettersor any attachments. A minority provided acovering letter and only 3 included a detailedattachment. Only 24 of the 67 respondents

stated that they had undertaken a formal reviewof their IT systems. Another 21 state they planto during 1997 or 1998 (with one as late as1999).

The remaining respondents are not planning anassessment or are leaving this assessment solelyin the hands of external providers.

Those organisations that have undertaken anassessment of the impact have typicallycompleted this assessment in late 97 and expectto resolve any issues discovered in the reviewby on average mid 1998. On average thoseorganisations who had not undertaken anassessment of their systems planned to do so inthe next 12 months.

Other information from the survey suggest thatthese organisations are currently unable toidentify the expense associated with makingtheir systems Year 2000 compliant. Manyrespondents left this question blank. Some ofthe accountancy partnerships mentioned theywill be upgrading their hardware and softwareprior to 2000 and thus these expenses will beabsorbed by their general IT budget.

Larger organisations

There were 70 responses listed in this category.48 respondents stated they had completed or arecurrently undertaking a formal review of theirIT systems and a further 18 provided a datestating when they would start. 28 of therespondents provided detailed attachmentswhile others provided brief comments in either acovering letter or on the questionnaire.Generally speaking this group of largerrespondents provided a far superior detailedresponse to the smaller respondents. The largebanks and insurance companies who replied tothe survey typically have a dedicated Year 2000team reporting to management on a regularbasis. Budgets for solving the Year 2000problem range from no budget to $250 million.

Findings

It would appear from the survey results thatmany organisations in the superannuationindustry are not taking the Year 2000 issueseriously. Many of the questionnaires containblanks. Many respondents when given theopportunity to expand their answer in anattachment did not do so. A number oforganisations seem to be ignoring the problemand the low response rate to the survey is notencouraging. Interestingly two of the smalleragencies claim to have assessed their systems asearly as 1994 when there was scant informationon this problem.

Some of the agencies both small and large willneed to clarify their target dates for reachingyear 2000 compliance. Some dates forrectification were quoted as late 1999 or eventhe year 2000 itself. This is too late ascomputing resources are expected to be in shortsupply at this time. Additionally, the nature ofsuperannuation may require rectification prior to2000. For example benefit quotations may beissued 12 months before a member’s exit date in2000.

Conclusion

Despite wide spread media coverage of the year2000 issue the poor response to theCommission’s questionnaire suggests manyparticipants in the superannuation industry arenot taking this issue seriously enough. It wouldappear small to medium organisations inparticular are not addressing this issue in aproactive manner and are preferring to wait andsee what happens closer to the date. This isneither acceptable nor a proper demonstrationof trustee responsibilities.

Infrastructure investment

One of the strongest growing asset classes for superannuation funds is infrastructure investment. Inthis article we provide a recent snapshot of this important investment class.

According to a survey recently published inSuper Review magazine the Australian andNew Zealand infrastructure investmentmanagement market now totals A$5.3 billion.

This infrastructure investment representsaround two per cent of super assets comparedto an ASFA estimate of around five per cent,and an AIMA estimate for the end ofDecember 1996 of 0.3 per cent. Determiningthe levels of infrastructure investment iscomplicated however by the difficultiesinvestment managers' accounting systems havein differentiating specific infrastructure equityand debt investments from other moretraditional equity and debt investments.

The infrastructure investment market isextremely concentrated as the largest manageris responsible for nearly half of this totalinfrastructure investment with $2.2 billioninvested in infrastructure. See figure 1. In1996-97 infrastructure investment experiencedrapid growth increasing by $1.5 billion or 39per cent, nearly twice as fast as the overallsuperannuation industry.

Infrastructure investment is becoming moreattractive to superannuation fund trustees andother investors. For example, mostinfrastructure investment managersexperienced growth in assets undermanagement of around 100 per cent.Emphasising this trend some institutions havereported making significant provision forfurther infrastructure investment asopportunities arise.

Most infrastructure investment is made viaequity investment. The value of infrastructureequity is $3.6 billion (68 per cent of totalinvestment) compared with only $1.6 billioninvested in infrastructure debt. Some equityinvestment is in the form of direct investmentin infrastructure projects. Many investmentinstitutions limit themselves to providing lessthan 50 per cent of the investment capital inany particular infrastructure project.

All companies prefer to invest predominantlyin infrastructure projects in local markets,however infrastructure investments have alsobeen made in New Zealand and Asia.

Figure 1: Infrastructure investment by investment managers

0

500

1000

1500

2000

2500

AMP Hastings HRLMorrison

MacquarieBank

CFS Axiom DevelopmentAustralia

LendLease

MercantileMutual

Investment before 1996-97

Investment in 1996-97

Source: Super Review September 1997.

Using investment managers - the US experience

While comparisons between the use of investment managers by Australian superannuationfunds and their US equivalents (pension funds) are influenced by the large differences in boththe size and maturity of their respective investment management markets, the US experience

may nonetheless provide some possible directions for the Australian industry.

The manner in which superannuation fundtrustees approach the investment of their fund’sassets and the relationship that they build withtheir investment manager(s) has an importantbearing upon the investment returns achievedby the fund. Recent surveys conducted inAustralia by the Association of Superannuationof Australia (ASFA) on fund investmentpatterns1 and in United States (US) byPensionforum2 have focused on thistrustee/investment manager relationship. Thisarticle analyses some of the major findings ofthese two surveys.

US pension funds on average use around twicethe number of investment managers used byAustralian superannuation funds, at around 10and five investment managers respectively.Moreover over two thirds of Australian superfunds use between one and five managers,compared to around 27 per cent of US pensionfunds.

This result most likely reflects the greateraverage size of the US pension funds, whichleads in turn to improved access to wholesaleinvestment management markets and greaterability to diversify among managers. In fact,use of investment managers by Australian

funds varies dramatically with the level ofassets under management (AUM). Forexample, Australian funds with less than $70million AUM on average use around threeinvestment managers, compared to funds withmore than $70 million AUM who on averageuse around seven managers.

However, perhaps of more interest to theAustralian industry is that over the last threeyears US pension funds have been increasingthe number of investment managers that theyuse. During this time more pension fundsshifted to using more than five investmentmanagers, with the result that the averagenumber of investment managers has increasedfrom nine to ten. See figure 1.

This trend for US pension funds shows no signof abating in the near future, as twice as manyfunds indicated that they intend to increase thenumber of investment managers they use thanfunds that indicated they intend to reduce thenumber. For example, 30 per cent of US fundsindicated they intend increasing their numberof investment managers by the year 2000 whileonly 15 per cent indicated that they intendedreducing the number. This result clearly

Figure 1: Use of investment managers

0%

10%

20%

30%

40%

50%

60%

70%

1 to 5 6 to 10 11 to 20 more than 20

U.S. (3 years ago)

U.S. (now)

Australia

Source: ASFA Investment Patterns survey, July 1997 (Australia) Pensionforum, Institutional Investor, October 1997 (US)

1. ASFA Research and Statistics committee Investment Patterns survey, July 1997.

2. Pension forum results published in Institutional Investor, October 1997.

indicates a strong and increasing preference byUS pension funds for having a number ofspecialist investment managers, for example aspecialist equity manager, a specialist fixedinterest manger and so on, in preference to a‘macromanager’.

A macromanager is a single large investmentmanager who provides investment managementservices across a range of asset classes. Fundstaking the macromanager approach may beable to achieve lower investment managementfees and improved service, however themacromanager may not have as high expertiseas specialist managers across all the assetclasses that it offers.

Manager review and monitoring

Regardless of the approach taken by the fund,there is constant pressure on investmentmanagers to perform. Reflecting these marketpressures, over each of the past two yearsnearly 50 per cent of US pension funds haveterminated at least one of their investmentmanager contracts.

In Australia as well, investment managers areclosely monitored by their superannuation fundclients, with 58 per cent of super fundsindicating that they monitor their managersconstantly. Issues of special interest tosuperannuation funds include poorperformance during a reporting period or asignificant shift in staff away from aninvestment manager.

Defined benefit funds tend to review theirinvestment managers less than frequently thanaccumulation funds. For example definedbenefit fund often review their investmentmanagers review their managers at the sametime as they have their actuarial review(usually every three years).

Once a new investment manager has beenappointed, over 75 per cent of Australian fundsindicated that they would have reviewed thenew manager’s performance within 12 months,with this figure rising to 98 per cent within twoyears.

The vast majority of Australian super funds(around 80 per cent) measure the performanceof their investment manager against anindividual preset benchmark. Otherperformance measures used by funds includecomparisons with other investment mangers(used by 58 per cent of funds) and comparisonwith CPI plus some variable percentage return(used by 33 per cent of funds).

While poor performance will most likely leadto a contract with an investment manager beingterminated, it is not the sole reason funds usewhen deciding to terminate a manager. Whilepoor performance accounted for some 57 percent of contract terminations by US funds,other significant reasons included arealignment of asset allocation by the fund, achange in ownership of the manager and achange in portfolio management talent (staff)at the manager. See table 1.

Allocating investment returns

Once the investment managers have producedan overall return for a super fund, it is then amatter for the fund trustees to determine theinvestment return to be credited to eachmember. Funds may credit the full return tomembers, or else may use the actual return tocalculate a crediting rate taking into accountthe funds' averaging or reserving strategy.These strategies are used by funds to smooththe returns credited to members, by reducingthe fluctuations between high and lowperforming years.

Table 1: Reasons for terminating investment manager contracts

Termination reason Proportion of terminations

Poor performance 57%

Realignment of asset allocation 18%

Change in portfolio management talent 11%

Change in ownership at manager 7%

Other 7%

Source: Pensionforum, Institutional Investor, October 1997

Around 61 per cent of large Australian superfunds utilise some form of averaging orreserving strategy when determining thecrediting rate to apply to their members. Seefigure 2. These funds tend to be predominantlydefined benefit and combination (funds with

both defined benefit and accumulationelements.) The majority of funds (55 per cent)that use averaging methods do so over threeyears, however some funds indicate theyaverage over five years.

Figure 2: Method of determining the fund’s crediting rate

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

Average Full interest Use reserves

Source: ASFA Investment Patterns survey, July 1997

Focus on corporate superannuationAfter the introduction of the SIS legislation it appeared that many corporate funds reviewedtheir overall position in the market and took the opportunity to restructure. In this article weprofile the corporate superannuation market and examine the important consolidation trends

that the sector is experiencing.

Corporate superannuation funds are fundssponsored by a single non-governmentemployer or group of related employers.Corporate funds, usually being standardemployer-sponsored funds, are legal entities intheir own right. This is in contrast to employersponsored superannuation arrangements thatmake use of mastertrusts. In these cases eachemployer arrangement usually operates as asub-fund within the mastertrust, while themastertrust itself is considered to be a retailfund.

Traditionally, employers established corporatefunds as a means of providing additionalbenefits (and 'golden handcuffs') to selectedemployees and as a means of differentiatingthemselves from other employers in the labourmarketplace.

However, the advent of award superannuation,followed by the introduction of thesuperannuation guarantee (SG) has madesuperannuation a standard entitlement forvirtually all workers. This has led manyemployers to reconsider the commercial costsand benefits of providing employees withsuperannuation over and above the compulsoryminimum.

After the introduction of the SIS legislation itappeared that many corporate funds had

reviewed their overall position in the marketand taken the opportunity to restructure. Thisprocess is likely to accelerate with theintroduction of member choice of fund.Nonetheless, for a number of employers theprovision of quite generous superannuationbenefits remains an integral part of theiroverall remuneration packaging strategy.

Corporate fund profile

Corporate funds are by far the most commontype of large superannuation fund, representingaround 85 per cent of all non-excluded funds;accounting for over $63 billion, or 24 per centof all superannuation fund assets, at June 1997and of all superannuation accounts. Previousanalysis published by the ISC indicates thataround three per cent of private sectorbusinesses employing more than fiveemployees operate their own superannuationfund.

The most recent census of corporatesuperannuation funds available is that providedby the 1995-96 ISC Annual Returns. Thefollowing profiles are based upon an analysisof that information.

At June 1996 there were around 4 670corporate superannuation funds, managingover $55 billion on behalf of their members.Concentration in the corporate funds is not as

Figure 1: Corporate superannuation fund profile 1995-96

0%5%

10%15%20%25%30%35%40%45%50%

5 to 50 51 to 200 201 to 500 more than 500

Member range

fund

s

0%

10%

20%

30%

40%

50%

60%

70%

assets

accumulation (LHS)

defined benefit (LHS)

accumulation (RHS)

defined benefit (RHS)

pronounced as for superannuation funds as awhole. The majority of funds (around 60 percent) are in the five to 50 accounts range andthe majority of assets (77 per cent) are withfunds in the 500 and above accounts range.Even though the majority of corporate funds(70 per cent) are accumulation funds, themajority of corporate fund assets (79 per cent)are managed as defined benefit funds1 SeeFigure 1.

Corporate accumulation funds havesignificantly different per account profilescompared to corporate defined benefit funds.For example, on average, defined benefit fund

assets per account and expenses per accountare around twice their accumulation fundequivalents within each member range2. Thelevel of expenses per account also falls asmembership grows for both accumulation anddefined benefit funds, most likely reflecting theeconomies of scale available to funds withlarger memberships. Anecdotal evidencesuggests the greater level of assets per accountexperienced by both accumulation and definedbenefit funds with the smallest membershipmost likely reflects the corporate executiveschemes that are contained in this category.See Table 1.

Table 1: Corporate fund assets and expenses 1995-96

Assets per account ($) Expenses per account ($pa)

Account range Accumulation Defined benefit Accumulation Defined benefit

5 to 50 50,771 116,562 340 996

51 to 200 26,465 65,760 249 518

201 to 500 24,045 53,329 197 365

more than 500 13,210 57,177 85 250

Overall 17,073 58,111 122 287

In terms of asset size, only four per cent ofcorporate funds have more than $50 millionassets under management (AUM), with thesefunds managing around 68 per cent ofcorporate sector assets. The vast majority ofcorporate funds (85 per cent) have less than$10 million AUM, including the 49 per cent(around 2 290) of corporate funds that haveless than $1 million AUM. These smallestfunds collectively manage less than $1 billion(around one per cent) of all corporate sectorassets.

Consolidation trendsDuring 1995-96 around 15 per cent ofcorporate funds wound up, a rate only slightlybelow the 1994-95 wind up rate of 17 per cent.While during 1994-95, many continuing fundsundertook significant restructuring as a resultof the introduction of the SIS legislation, oftensplitting into a number of excluded funds, in1995-96 the incidence of restructuring wassignificantly reduced. See figure 2.

Figure 2: Corporate sector dynamics 1995-96 (1994-95)

June 1995

June 1996

Corporate funds

Wound up Continuing Converted to an excluded fund

15% (17% ) 80% (49% ) 5% (34% )

1. Defined benefit funds and combination funds (ie. funds containing both accumulation and definedbenefit elements) are jointly described as defined benefit funds in this article.

2. As defined benefit funds are unallocated and do not technically have an account for each memberthese figures are notional and indicative only.

This result suggests that any fund restructuringbrought about by a perceived regulatoryarbitrage under SIS dissipated during 1995-96.However, some industry commentators havesuggested that when, as foreshadowed in theTreasurer's Statement on Financial SystemReform of 2 September 1997, the regulatorysupervision of excluded funds is transferred tothe Australian Taxation Office (ATO) this maypossibly cause a further round of restructuringwithin the industry.

Previous analysis published by the ISCindicated that transfers from elsewhere withinthe superannuation system accounted for nearly60 per cent of deposits into mastertrusts,compared with around 15 percent of depositsinto industry funds. This result stronglyindicates that mastertrusts rather than industryfunds received the largest share of the assetsbecoming available for management elsewherein the superannuation system as a result ofcorporate fund rationalisation.

A further factor influencing the decrease incorporate fund numbers may have beenthrough larger employers rationalising theirvarious superannuation arrangements. Forexample, this restructuring may have occurredwhere, through mergers and/or acquisitions, anemployer that sponsored several funds withsimilar member profiles may have taken theopportunity to rationalise these arrangementsinto a single fund.

In overall terms the number of corporate fundsfell from around 5 090 at June 1995 to around4 670 at June 1996, a decrease of slightly morethan eight per cent. The difference between thewind up rate of 15 per cent and the decrease inoverall corporate fund numbers of only eight

per cent is explained by the number of newcorporate funds that established during 1995-96 (representing nearly four per cent of June1996 corporate funds) and by funds increasingtheir membership size and moving from beingan excluded fund at June 1995 to being acorporate fund at June 1996 (representing fourand a half per cent of June 1996 funds).

The account range (i.e. membership size)profile of the newly establishing corporatefunds is similar to the continuing corporatefund profile. For example, 70 per cent of thecorporate funds established during 1995-96had between five and 50 members. Theestablishment of funds with low numbers ofaccounts, and most likely with relatively highoperating costs, suggests strict commercialconsiderations are not the sole determinant foran employer in the decision of whether or notto operate a superannuation fund as a distinctlegal entity. Moreover, a small number of thesenew funds may be funds that are rolling outfrom a mastertrust, perhaps as a result of adesire by the employer to have greater controlover their superannuation arrangements thanthey were able to within the mastertrust.

Retail and corporate consolidation differences

While much of the focus on fundrationalisation has been on the corporatesector, the reality is that the number of retailfunds has been decreasing at an even fasterrate. For example, while the overall number ofcorporate funds decreased by eight per centduring 1995-96, the number of retail fundsdecreased by more than 26 per cent over thesame period. However the reasons behind theconsolidation of retail funds are very differentto those for corporate funds.

-$4,000

-$2,000

$0

$2 ,000

$4 ,000

$6 ,000

$8 ,000

95-96 96-97 95-96 96-97 95-96 96-97

$ m

illio

n

N e t c o n tr ib u t io n s

N e t t r a n s fe rs

N e t d e p o s it s

C o rp o rate Re ta il In d u s try

Figure 3: Superannuation financial flows, 1995-96 to 1996-97

Retail funds regulated under the SIS legislationare often legal umbrella structures that sit overthe top of a number of individual retailproducts. Members joining these retail fundsdo so indirectly through purchasing a particularproduct that is considered to be part of thelegal fund. In the case of many of the largerproviders of retail superannuation products,particularly life offices, their wide range ofproducts had in the past been covered by anumber of legal umbrella funds. Moreover,these funds were usually all managed by thesame Approved Trustee company, usually asubsidiary of the product provider. Therationalisation of retail funds hasoverwhelmingly been due to these large retailsuperannuation product providersconsolidating their legal structures so that alltheir products now reside within a muchsmaller number of funds, in some cases justone. In these cases, the number of individualproducts usually remains unchanged despitethese new legal arrangements. This means thatwhile the number of retail funds has decreaseddramatically, the number of retail productsoffered has not.

Anecdotal evidence suggests that the industryfund sector may also begin to experience arationalisation in fund numbers over the shortto medium term.

Figure 3 shows that during 1996-97 the netoutflows from the corporate sector increasedby 68 per cent, while in contrast the net inflows(deposits) to the retail sector and industry fundsector increased by 56 per cent and nine percent respectively over the same period.

Retail consolidation has therefore been verydifferent to the corporate sector consolidation,which has seen large outflows from the sectoras funds have wound up and employers haveoutsourced their superannuation arrangements.In other words, the retail consolidation hasbeen in fund numbers only, with no netcashflow component, while the corporateconsolidation has been characterised by a netcash outflow from the sector as well as adecrease in fund numbers.

Many of the figures published in the ISC Bulletin are derived from a combination of statistical estimates from the ISCQuarterly Survey of Superannuation (for large funds) and projected trends from prior year’s Annual Returns (for small

funds). In this article we assess the reliability of these estimates against the actual 1995-96 Annual Return figures.

The ISC Bulletin contains estimates for the structure ofthe superannuation industry based upon the resultsobtained from the ISC Quarterly Survey ofSuperannuation (the Survey) and prior year statutoryAnnual Returns. Importantly, these estimates arepublished within three months of the end of the reportingperiod, so that June 1997 estimates for the structure of thesuperannuation industry are available in September 1997.

In contrast, up to date results from the audited AnnualReturns required by the ISC for the purpose ofsupervision do not become available until some 12 to 18months after the end of the reporting period (dependingon the type of fund). However, being auditedinformation, the Annual Return provides a usefulbenchmark against which to evaluate the validity of theestimates published in the ISC Bulletin.

Reliability of the ISC Bulletin

The close correlation between the previously publishedestimates and the results obtained from the 1995-96Annual Return confirm the overall robustness of themethodology used to produce the superannuation industryestimates published in the ISC Bulletin. Nonetheless,these derived figures appear to underestimate the Annualreturn results across the superannuation sector with theexception of public sector funds and assets of thecorporate sector.

Previously published figures appear to haveunderestimated the Annual Return result by around twoper cent for total assets and one per cent for totalmembers. See table 1. The difference between the publishedand Annual Return results for contributions is around sevenper cent. See table 2. This suggests the ISC measurementmethodology may in fact be rather conservative.

There is however greater variability in accuracy of estimatesbetween different fund types. For example, survey results forcorporate fund estimates of assets and contributions arealmost identical to the annual return figures. In contrast,excluded and retail funds assets were underestimated by fiveand nine per cent respectively in June 1996.

Contributions into retail funds were underestimated byseventeen per cent and contributions into excluded fundswere underestimated by twelve per cent in June 1996.

The implication of this is that the ISC Bulletin may beunderestimating contribution flows by several billion dollars.

Exempt public sector super funds

Due to strict solvency provisions in the SIS Act, not allconstitutionally protected funds are in a position to satisfy allSIS requirements. As a result, some of these public sectorfunds operate as "SIS exempt" schemes and consequentlythey do not lodge Annual Returns with the ISC. Nonetheless,many "SIS exempt" funds are included in the QuarterlySurvey of Superannuation.

For example, while the June 1996 survey includes 65 publicsector funds, 31 are actually "SIS exempt" and do not lodgeAnnual Returns with the ISC. As a result it is not possible totest the accuracy of survey results against an auditedbenchmark. SIS exempt schemes account for around 60 percent of public sector assets and 80 per cent of public sectorcontributions. The higher ratio of contributions to assetsamongst the SIS exempt schemes may reflect the unfundednature of some of these schemes.

Excluded funds

Since excluded funds (with less than five members) areoutside the scope of the Survey, their quarterly estimates

How reliable are the figures in the ISC Bulletin

Previously published estimatesJune 1996

Annual Return results1995-96

Fund type assets ($billion)

Corporate 55.4 55.2

Industry 14.2 14.2

Retail 59.7 62.6

Excluded 25.3 27.6

Public sector 59.4 59.5

Total assets 214 219.1

Total members (million) 16.2 16.3

Table 1 : Published estimates of assets and actual Annual Return results

are based upon previous years' Annual Returns and fundestablishment data. 'Per fund' asset and contribution ratiosare estimated on the basis of previous Annual Return data.Estimates are obtained by applying these ratios to fundnumbers.

There were close to 106 000 excluded funds that lodgedannual returns with the Commission in the financial year1995-96, from a total of 110 000 superannuation funds,ADFs and PSTs.

The underestimation of excluded fund assets ancontributions arises from the ISC's conservative estimationmethodology. While much of the increase in assets orcontributions is driven by the increase in the numbers ofexcluded funds, ISC makes the assumption that 10 per centfunds are discontinued each year.

Moreover, investment income for excluded funds is alsoconservatively estimated. For example, one third of excludedfund assets are assumed to be held as cash deposits and afurther 13 per cent of assets are estimated to increase only inline with the CP1. Holdings in equities or direct property areestimated to increase in line with national indicies.

Previously published estimatesJune 1996

Annual Return results1995-96

Fund type contributions ($billion)

Corporate 3.6 4.0

Industry 3.1 3.1

Retail 6.9 8.1

Excluded 3.8 4.2

Public sector 8.9 8.8

Total contributions 26.3 28.2

Table 2 Published estimates of contributions and actual Annual Return results

Table 3: Survey EstimatesPublic Sector superfunds, June 1996

Assets Contributions

($billion)

SIS Exempt 35.4 6.8

SIS Regulated 24.1 2.2

Total 59.4 8.9

Life insurance asset allocation profilesIn 1997 the ISC introduced a new statutory reporting regime for life insurance companies.In this article we report on the June 1997 asset allocation profile, particularly focussing

upon the structure of overseas asset holdings.

Following the introduction of the LifeInsurance Act 1995, the ISC has beenworking with life offices to redevelop andimprove the statutory returns that companieslodge with the ISC and to also improve thequality of statistics available from thesereturns.

While these new data collections have beenunderway for some time now and have beenincorporated into the processes by which theISC prudentially monitors the life insuranceindustry, only a small amount of this newinformation has to date been widelypublished.

In June 1997, there were 51 companiesoperating in the Australian life insurancemarket accounting $139.7 billion in assetsunder management. Superannuation assetsmake up $111 billion, or 76 per cent of theseassets. Viewed from another perspective, lifeinsurance companies hold 38 per cent of allsuperannuation assets.

Asset allocation

The majority of life office assets are held as

either equities or fixed interest securities, with39 per cent held in equities and 37 per centheld in fixed interest. Direct propertyaccounts for nine per cent of assets, loansaccount for six per cent and cash accounts forfive per cent. The remaining four per cent isheld in other miscellaneous asset classes.

Further analysis also reveals that largercompanies tend to be more aggressiveinvestors with higher equity weightings thansmaller companies. Conversely, smallercompanies tend to have higher weightingstowards interest bearing securities. Largercompanies also have property weighting up tothree times as high as small companies,possibly reflecting their ability to constructmore balanced portfolios through moresophisticated and diversified investmentmanagement operations. See figure 1.

The proportionally larger fixed interestweighting and lower equity weighting of thesmaller companies may be due them having toprotect their solvency ratios by limitingexposure to the more volatile higher riskassets.

Figure 1: Asset allocation trends and assets under management (AUM)

0%

10%

20%

30%

40%

50%

60%

1 10 100 1,000 10,000 100,000 1,000,000

Property

Fixed Interest

Equities

Life office AUM ($millions)

Overseas asset holdings

Life office investments in Australia at June1997 amounted to $117 billion, or 84 per centof total statutory fund assets.

Nearly three quarters, 73 per cent, of the $23billion that is invested overseas by life officesin Australia is held in equities. The remainingportion of overseas assets are heldpredominantly in interest bearing securities,accounting for 19 per cent of overseas assets.Directly held property accounts for only threeper cent, and the remaining five per cent ofoverseas assets is held in cash, loans andmiscellaneous classes.

The proportion of overseas assets reported asbeing held in equities is also very similar tofigures reported by the Australian InvestmentManagers Association in their December 1996Funds Under Management Report, whichrevealed that 80 per cent of overseasinvestment are held in equities.

The significant equity weighting for overseasassets is more than twice as high for the equityweighting of domestic assets. See figure 2.

The overseas holdings of life offices aredistributed across Europe with $6.1 billioninvested, the United States and Canada with

$5.9 billion, the Pacific and New Zealand with$2.9 billion, the Pacific and New Zealand with$2.9 billion, and South East Asia with$2.3 billion. However, the value of life officestatutory funds' "minor" overseas holdings -which are held in other portfolios -nonetheless aggregates to $5.4 billion, or 24per cent of all overseas holdings. Undercurrent prudential reporting rules, the countrywhere the minor holding is placed, if the valueis less than one per cent of the statutory fund,is not required to be separately identified.

More importantly, the value of investmentexposure to overseas countries, especiallySouth East Asia, should not be seenexclusively in terms of direct investments inthose countries, as many of the investments inAustralian companies are made on the basis oftheir export exposure to those economies.

Thus the real exposure of Australian policyholders to South East Asian economic eventsis likely to be somewhat higher than is initiallysuggested through only two per cent of theirassets being invested there. However, thefinancial strength of Australian life offices issuch that this does not raise any concerns.

0%

10%

20%

30%

40%

50%

60%

70%

80%

Property Equities Fixed Interest Other

AustralianOverseas

Figure 2: Domestic and Australian asset allocation

General Insurance Industry UpdateIn recetnly released December 1996 "Selected Statistics on the General Insurance Industry" the ISC General InsuranceGroup reported on the major trends impacting this important sector of the financial system. This article is a summary of

their key findings.

General insurance business (i.e. insurance other than lifeand health insurance) was written in Australia by 170private sector insurers and 16 public sector insurers as at30 June 1997.

A key indicator of general insurance activity is the rateof growth of direct premiums. Direct premium is theamount paid by business and consumers for theirinsurance cover. It includes stamp duties and otherGovernment charges, as well as intermediarycommissions.

The rate of growth of total direct premiums over the past5 years has exceeded the growth in GDP by an averageannual factor of 1.35 percentage points. Growth inpremium revenue nonetheless varies significantlybetween the different classes of general insurancebusiness.

Growth in compulsory classes of business for examplewas largely due to CTP Motor Vehicle insurance whichaccounted for almost half of the total increase inpremium revenue. The increase in CTP premiumsreflects rate increases necessary to address the significantunderwriting losses in the CTP class of business.

The largest insurance classes, motor vehicle anddomestic home insurance, which together account foralmost 40 per cent of private sector direct premium havecontinued to grow strongly. The rate of growth ofdomestic classes is greater than for commercial classes,which may reflect a trend towards increasing use of non-insurance risk management strategies and the degree ofprice competition in the commercial area.

There has also been a continuation of the movement ofbusiness from public sector to private sector generalinsurers. See figure 1. During 1996, total private sector

direct premiums increased by $954 million, double therate of growth of the previous year.

Underwriting PerformanceGeneral insurance business performance can bemeasured by the underwriting result, that is, premiumsless reinsurance, claims and other underwriting expenseshave made underwriting losses throughout the 1990s todate, even though there was a slight improvement in theunderwriting loss in 1996, down to $807 million from$890 million in 1995 ( an improvement of nearly 10 percent).

In the case of direct underwriters, insurers who deal withthe public at large, the underwriting deficit fell to$799 million in 1996 from a deficit of $987 million in1995. This improvement in underwriting performancecan be attributed to an increase in the CTP MotorVehicle class of business where premium revenueincreased from $1.1 billion in 1995 to $1.5 in 1996,while claims expenses remained constant at $1.8 billion.

Underwriting results vary widely across different classesof business with the compulsory classes (employerliability and CTP motor vehicle) as well as domesticmotor vehicle insurance continuing to make the largestlosses. In contrast, some niche classes, for example,Loan, Mortgage and Lease insurance have generallymade underwriting profits in recent years.

Nonetheless, in 1996, Loan, Mortgage and Leaseinsurance experienced the greatest deterioration inperformance by a particular class of business as the netclaims expense increased by 178 per cent, while netpremiums increased by only 14 per cent. Consequently,the underwriting result declined from a profit of $14million in 1995 to a loss of $6 million in 1996.

60

80

100

120

140

160

180

1992 1993 1994 1995 1996

1992

=100

Private Sector Public Sector

Total GDP Index

Figure 1: Premium Growth

Investment Returns and ProfitabilityDuring the 1990s private sector general insurers’underwriting losses have been more than offset byinvestment income, thus enabling insurers to return overallprofits while losing money on their insurance business.

Overall profits are sensitive to fluctuations in investmentincome such that the industry has been generally dependentover the last five years on investment returns forprofitability. See figure 2.

Net profits after tax fell sharply in 1994 when bond marketlosses caused significant decline in investment income. Since then, improved investment returns have led to animprovement in general insurers' profitability.

Reflecting intense price competition and strong investmentskills, general insurers have not exited the industry or raisedpremiums sufficiently to ensure that underwriting losses arenot incurred. This suggests there is limited scope forunilaterally raising premiums, so that investment income islikely to continue to play a crucial role in maintaininggeneral insurer’s profitability.

Income from invested premiums can also be consideredpart of the insurer’s risk business and so it is thereforeuseful, for management and other decision makingpurposes, to apportion the investment income betweenpolicyholders’ and shareholders’ funds.

Using this approach, this notional allocation of investmentincome to policyholders' funds converts an underwritingloss in 1996 of $807million to a profit of $1,015million.Similarly for 1995, an underwriting loss of $890 millionconverts to a profit of $939 million.

These results relate to premium revenue of $14.3 billion in

1996 and $12.8 billion in 1995. The notional return onshareholders' funds, on the other hand, represents a returnof $1.3 billion in 1996 and $1.2 billion in 1995 on netassets of $10.2 billion and $9.3 billion respectively. Seetable 1.

Assets and LiabilitiesTotal assets of private sector insurers increased from $23billion to $38 billion during the last five years, representingannual growth of nearly 11 per cent. This compares to a sixper cent annual growth in premiums over the same period.Net assets have also increased over the same periodbecause total liabilities have only grown at an annual rateof nine per cent, to $10.2 billion in December 1996.

Around one third of assets are non-income producing suchas unpaid premiums and reinsurance recoverable onoutstanding claims. Investment assets of private sectorinsurers continue to be well spread over the differentcategories of investments.

General insurers seem to be more conservative in theirinvestment strategies than superannuation funds and lifeoffices which have a noticeably lower weighting in debtsecurities and higher weighting in equities. This is primarilydue to a larger portfolio of shorter term liability risks whichgenerally necessitates a more liquid investment portfoliorelative to life insurers and superannuation funds.

For example, the share of equities and unit trusts in totalinvestments of super funds and life companies is 39 and 34per cent respectively compared to only 25 per cent forprivate general insurers. See figure 3. Debt securities were44 per cent of private sector general insurers' totalinvestment assets and have also. been the most importantsource of investment revenue. Consequently investmentreturns for general insurers could be expected on average tobe less variable, though lower, over time.

Figure 2: Trends in Profitability

-1500

-1000

-500

0

500

1000

1500

2000

2500

1992 1993 1994 1995 1996

$ m

illio

n

Underwriting Result Investment Income Net Profit after tax

SolvencyThe Insurance Act requires authorised general insurers tomaintain an excess of assets over liabilities of not less than$2 million, 20 per cent of premium income or 15 per centof the outstanding claims provision, whichever is thegreater. For most insurers, the required solvency margin isbased on 20 per cent of premium income which for thegeneral insurance industry was $ 3.1 billion at the end of 1996.

In aggregate, the industry comfortably meets the statutoryrequirement with a solvency surplus of $3.1 billion. Thissurplus is the largest since industry wide solvency marginswere first published in June 1994. However changes in themethodology used to compile industry wide solvencymeasures makes it difficult to establish a trend.

Market ConcentrationThe general insurance industry is still only moderatelyconcentrated but the level of concentration continues torise. In 1996, 75 per cent of direct premiums were writtenby the 20 largest insurers, compared to 77 per cent in 1995and 74 per cent in 1994.

Most classes of general insurance business are dominatedby underwriters who specialise in that particular class ofbusiness and therefore the concentration within individualclasses of business is higher than the average concentrationacross all industries.

For example the share of direct premiums written by the top10 insurers in each class of business is greater than 60 percent compared to only 52 per cent for all classes ofbusiness.

This characteristic of the market can also seen in four of thesmallest classes of general insurance business: TradeCredit; Loan Mortgage and Lease; Extended Warranty andConsumer Credit, where the ten largest insurers, wrote 100per cent of the business in 1996.

Product DistributionGeneral insurance business in Australia is written both'over the counter' (without the use of intermediaries) andalso through intermediaries acting as agent of the insurer,or the insured. Insurance brokers, acting as agent of theinsured, are required to be registered under the Insurance(Agents and Brokers) Ac 1984.

There are currently around 1,020 registered generalinsurance brokers who play a significant role in the sale ofgeneral insurance policies. During the last five yearsbrokers accounted for between 43 per cent and 51 per centof annual direct premiums collected.

Brokers however tend to play a more minor role indomestic insurance classes, as these products are often soldwithout an intermediary.

Table 1: General Insurers' Profitability 1992-96 Actual 1992 1993 1994 1995 1996Underwriting Result -393 -616 -216 -890 -807Investment Income 1,452 2,047 558 2116 2,312 Net Profit after tax 694 870 175 852 811 NotionalInvestment Income for Policy holders 655 790 240 939 1,015Insurance' Surplus (Deficit) 262 174 24 49 208Shareholders Investment Income 797 1,257 318 1,177 1,297

Note: the 'actual results do not add through because certain items (e.g. administrative expenses) are not shown.

Figure 3 : Share of major asset classes

0%

5%

1 0 %

1 5 %

2 0 %

2 5 %

3 0 %

3 5 %

4 0 %

4 5 %

di rec t p rope r t y d e b t s e c u r i t i e s equ i t i es & un i t t r us t s d e p o s i t s , c a s h &l o a n s

o t h e r a s s e t s ou t s i de Aus t ra l i a

Per

cent

age

shar

e of

inve

stm

ent a

sset

s

P r i v a t e s e c t o r g e n e r a l i n s u r e r s ( t o t a l $ 2 7 b i l l i o n )Supe rannua t i on ( t o ta l $230 b i l l i on )

Superannuation survey highlights - September 1997

Main features• Τotal superannuation assets had reached

$316.7 billion by end September 1997,representing growth of 4% during thequarter, or 20% during the year endedSeptember 1997.

- note that the effects of the October slump in capital markets are not reflected in these figures.

• Capital market performance during theSeptember quarter caused net earnings tobe the main component of growth,accounting for 76% of net growth. Netdeposits accounted for only 24% of thegrowth.

• Contributions this financial year were up11% compared to the previous 12 months,increasing from $27.3 billion to$30.3 billion (this represents a very slightdecrease in the growth of contributionscompared to previously reported figures).

• Even after discounting the rapidly growingexcluded fund sector, contribution growthfor large funds is still 10% per annum.

• The strongest growth came from membercontributions, increasing by 17% over theprevious year to $10.4 billion. Employercontributions increased by 8% to $19.9billion.

Industry structureThe assets managed by small self-managedfunds (ie, excluded funds with less than 5members) grew fastest during the year endedSeptember 1997 increasing by 36% ($9.7 billion). This was closely followed byindustry funds which grew by 32% ($5.1billion) during the last year.

Corporate fund assets grew by only 15%, or$8.5 billion during the year. Retail assets grewby 23% ($14.6 billion) and public sector assetsgrew by 24% ($14.3 billion).

Retail funds currently hold around 24%($77.2 billion) of total superannuation assets,public sector funds hold 24% ($74.9 billion),corporate funds 21% ($65.0 billion), excludedfunds 12% ($36.9 billion), and industry funds7% ($21.2 billion).

The excluded fund market segment share hasgrown from 10% to 12% during the year ended

September 1997. However most marketsegment shares remained relatively stableexcept for a decline in the market segmentwhich represents annuity products, fundreserves and unallocated profits of life officestatutory funds. The proportion of thesuperannuation industry represented by these‘balance of statutory fund’ assets has reducedto 13%.

Contributions and benefitsDuring the September quarter, employerscontributed slightly over $5.1 billion intosuperannuation, up 9.0% on the 1996September quarter. In contrast, the $2.7 billionemployees contributed into superannuationduring the same period was up 15.5%.

The contributions into small self-managedfunds were 17.6% higher during the year endedSeptember 1997 than the year endedSeptember 1996. Growth in net inflows tothese funds was 44.4% higher than in theprevious 12 months, being largely fuelled bythe strong growth in the number of excludedfunds, eg. the number of excluded fundsincreased to 163,173 by September, up 4%during the quarter.

Reflecting very strong consolidation incorporate and retail fund numbers this quarter,inward transfers accounted for 65% of allmoney deposited into superannuation duringthe September quarter, which is considerablyhigher than the normal average of around 36%.

Lump sums, excluding outward transfers,accounted for 81% ($4.6 billion) of thebenefits paid during the September quarter.The remaining 19% ($1.1 billion) of benefitswere paid as pensions. Outward transfers, forreasons referred to previously, accounted for70% of all fund withdrawals during theSeptember quarter.

A major contributor to the rapid growth intransfers between super funds has been changesin the structure of public sector super funds.However, this strong growth in public sectortransfers overwhelmingly reflects significantrestructuring and rationalisation, in particularin relation to governments seeking to managetheir potential defined benefit liabilities.

Benefit payments, excluding transfers, duringthe year ended September 1997 were up by11% compared to the previous 12 months.

Despite the higher growth rate of benefitpayments as compared to contributions, netcontributions (ie., contributions less benefits)were 7% higher for the year ended September1997 as compared to the previous 12 months.In other words, during this period more than$11 billion in net contributions flowed intosuperannuation than flowed out.

Manner of investmentAssets directly invested by trustees, showed thestrongest growth during the quarter, increasingby 6.4%. Assets placed with investmentmanagers which grew by 3.8%, while assetsinvested through the statutory funds of lifeoffices grew only 2.1% during the quarter.

Investment managers had 39.5%($125.1 billion) of total superannuation assetsat the end of September 1997, unchanged fromSeptember 1996. The share of directly investedsuperannuation assets increased marginally to24.2% ($76.8 billion) , with the statutory fundsof life offices falling marginally to 36.2%($114.8 billion).

Asset allocationThe share of superannuation assets investedoverseas declined slightly to 16.5%(46.6 billion) at the end of September 1997,largely due to the 1.5% appreciation of theAUD during the quarter.

Superannuation investment held in equities hascontinued to increase, growing by 4.3% duringthe September quarter. Since the ASXaccumulation index grew by only 2.4% in theSeptember quarter, it follows that there was asmall net inflow of around $1.6 billion into theequities markets by superannuation funds.Superannuation equity holdings overallincreased marginally to 28.9% of totalsuperannuation assets.

Unit trust holdings increased by 5.2%($1.5 billion) in the September quarter. Theyare now a record 10.8% of the total value ofsuperannuation assets.

Reflecting the decline in long term bond yieldsduring the September quarter, holdings of longterm debt securities increased by3.8% ($1.7 billion). The proportion ofsuperannuation assets held as long term debtsecurities has fallen slightly to 16.6%.

Holdings in short term debt securities rose by1.7% ($0.4 billion) during the Septemberquarter, partly reflecting falls in short termyields from 5.3% to 4.7%. The proportion of

superannuation assets held as short term debtsecurities declined slightly to 8.9%.

Holdings of cash, deposits and placementsgrew most strongly of all asset classesincreased by 4.5% in the September 1997quarter. This growth was well in excess of cashrates which declined below 5% during theSeptember quarter.

These movements would appear to indicatethat during the September quartersuperannuation funds were net sellers ofAustralian long and short term debt securities,but were net purchasers of Australian equities.

This result suggests that superannuation fundsare participating in profit taking in the strongbond markets and after some profit taking inthe equities markets in the previous quartersuperannuation funds are increasing their cashreserves and exposure to Australian securitieswhile generally maintaining existing weightingin foreign securities.

The value of assets held in direct propertycontinued to fall in the September quarter. Theshare on total superannuation assets fell to4.9% at the end of the September 1997 from6.4% a year earlier. Other investments accountfor around 4% of total superannuation savings.