Acts Interpretation Act 1901, s 15AB(1)(a)

Administrative Appeals Tribunal Act 1975, ss 35, 37, 44

Designs Act 2003

Income Tax Assessment Amendment (Research and Development) Act 1986 Act No 59 of 1988, ss 5(d), 5(e), 16

Income Tax Assessment Act 1936, ss 73B(1), 73B(4), 73B(1AAA), 73B(2A), 73B(2B), 73B(2BA), 73B(2C), 73B(10), 73B(11), 73B(12), 73B(14), 73B(33A), 73B(33C), 73B(34), 73BF(4), 73BM(4), 124ZF(1), 124ZG(2A)

Industry Research and Development Act 1986, ss 4(1), 39, 39AA, 39HG, 39HI, 39HH(1), 39J, 39JA(5B), 39K, 39KA, 39L, 39N, 39P, 39S(4), 39T(1)(a), 39T(4)

Part IIIA

Industry Research and Development Act 1986, Act No 82 of 1996, s 3, Sch 1, item 12

National Companies and Securities Commission Act 1979, s 36(2)

Olympic Insignia Protection Act 1987

Patents Act 1990, ss 3, 7(1), 9(a), 18(1), 18(1A), Sch 1

Tax Laws Amendment (Research and Development) Act 2011; Act No 93 of 2011, s 2, item 6 and s 3, Sch 3, Part 6, item 44, s3 and Sch 4

Taxation Administration Act 1953, ss 14ZZE, 14ZZJ

Taxation Laws Amendment Act (No 3) Act 1996, Act No.78 of 1996, s 3 and Sch 4, item 53

Taxation Laws Amendment Act (No 4) 1989, Act No. 167 of 1989, s34

Taxation Laws Amendment (Research and Development) Act 2001, Act No.170 of 2001, s 3, Sch 1, items 1, 2, 5, 6, 7

Tax Laws Amendment (2007 Measures No.5) Act 2007, s 3, Sch 11, item 1

Trade Marks Act 1995

Australian Trade Commission v F & F Asia Pty Ltd (1996) 69 FCR 252; 42 ALD 197

Attorney-General (NSW) v Mayas Pty Ltd (1988) 14 NSWLR 342

Avel Pty Ltd v Attorney-General for New South Wales (1987) 11 NSWLR 126

Azuko Pty Ltd v Old Digger Pty Ltd [2001] FCA 1079; (2001) 52 IPR 75; [2001] AIPC 91-741

Blacktown Workers' Club Ltd v O'Shannessy [2011] NSWCA 265; (2011) 183 LGERA 184

Brown v Commissioner of Taxation [2001] FCA 276; (2001) 47 ATR 143

Clements v Independent Indigenous Advisory Committee [2003] FCAFC 143; (2003) 131 FCR 28; 37 AAR 309

Confidential and Industry Research and Development Board [1997] AATA 67

Drake v Minister for Immigration and Ethnic Affairs (1979) 2 ALD 60; 24 ALR 577; 46 FLR 409

Heggie v Minda Incorporated [2006] SAIRComm 9

HR Products Pty Ltd v Collector of Customs [1990] FCA 152; (1990) 20 ALD 340

Hodgson v Imperial Tobacco Ltd [1998] EWCA Civ 224; [1998] 2 All ER 673

Industry Research Board v Coal & Allied Operations Pty Ltd [2000] FCA 979; (2000) 101 FCR 405; 44 ATR 541

Industry Research Board v Unisys Information Services Australia Ltd (1997) 77 FCR 552; 37 ATR 62

Jones v Dunkel [1959] HCA 8; (1985) 101 CLR 298; 32 ALJR 395

Kuswardana v Minister for Immigration and Ethnic Affairs [1981] FCA 66; (1981) 54 FLR 334; 35 ALR 186

McKeown v Repatriation Commission (1995) 39 ALD 30; 22 AAR 229

National Companies and Securities Commission v Bankers Trust Australia Ltd [1989] FCA 530; (1989) 24 FCR 217; 91 ALR 321

Pepsi Seven-Up Bottlers v Commissioner of Taxation (1995) 62 FCR 289; 132 ALR 632

Queensland v Allen [2011] QCA 311

Re Australian Federation of Construction Contractors; Ex parte Billing [1986] HCA 74; (1986) 68 ALR 416; 61 ALJR 37

Re Bolton; Ex parte Beane [1987] HCA 12; (1987) 162 CLR 514; 70 ALR 225; 61 ALJR 190

Re Charles IFE Pty Ltd and Industry Research and Development Board [1995] AATA 224; (1995) 39 ALD 635; 95 ATC 2149; [1995] AATA 224; 32 ATR 1226

Re Fermenter and Distiller and Industry Research and Development Board [2000] AATA 888; (2000) 45 ATR 1122

Re HZXD and Innovation Australia [2010] AATA 879

Re Mobil Oil Australia Limited and Industry Research and Development Board [1995] AATA 133; (1995) 95 ATC 2042; 30 ATR 1364

Re North Broken Hill Ltd and Industry Research and Development Board [1993] AATA 254; (1993) 30 ALD 200; 18 AAR 177; 26 ATR 1262

Re The Applicant and Industry Research and Development Board [1999] AATA 468; (1999) 55 ALD 784; 42 ATR 1116

Scott v Scott [1913] AC 417; [1911-1913] All ER 1

Secretary, Department of Social Security v Ekis (1998) 85 FCR 382; 52 ALD 246; 28 AAR 36

Transport Accident Commission v Bausch [1998] 4 VR 249; 13 VAR 61

Whitten v Falkiner [1915] HCA 38; (1915) 20 CLR 118

Chambers 21st Century Dictionary, 1999, reprinted 2004, Chambers

The Macquarie Dictionary, revised 3rd edition, 2001, The Macquarie Library Pty Ltd

Income Tax Assessment Amendment (Research and Development) Bill 1986, Second Reading Speech, Hansard, Senate, 2 June 1986 at 3152

Income Tax Assessment Amendment (Research and Development) Bill 1986, Explanatory Memorandum

Taxation Laws Amendment (Research and Development) Bill 2001, Explanatory Memorandum

DECISION AND REASONS FOR DECISION [2012] AATA 386

(1) decides to affirm the decision of the respondent dated

27 September 2007 and confirmed by a further decision dated 25 February 2010 to the effect that it is not satisfied that the applicant's activities are research and development activities as those activities are defined in s 73B(1) of the Income Tax Assessment Act 1936; and

(2) remits the matter to the respondent to give the Commissioner of Taxation a certificate to that effect under s 39L of the Industry Research and Development Act 1986.

Where an eligible company incurs research and development expenditure (other than contracted expenditure) greater than an aggregate of $20,000 during a year of income, the amount of that expenditure multiplied by 1.25 is allowable as a deduction from its assessable income in that year of income.[1] Among other criteria that it must meet, research and development expenditure must be expenditure in respect of research and development activities.

2. RACV Sales and Marketing Pty Ltd (RACV Sales),[2] which is a wholly owned subsidiary of RACV Holdings Pty Ltd, is an eligible company. It was registered under s 39J of the Industry Research and Development Act 1986 (IRD Act) in respect of research and development activities in each of the eight years of income in issue being 1998-1999 to 2005-2006 inclusive.[3] Although a pre-requisite for entitlement to a deduction, registration does not confer entitlement. The Research and Development Board (now Innovations Australia)[4] may or, if requested by the Commissioner for Taxation (Commissioner), must issue a certificate stating whether the activities are research and development activities. The Commissioner is bound by that certificate. The consequences of the certificate are not for us to decide. We note, though that if, as in this case, the Board decides that the activities, whether registered or not,[5] are not research and development activities, the Commissioner cannot allow a deduction under s 73B of ITAA36[6] for expenditure in relation to it in the year of income concerned.

3. In respect of each of those years, RACV Sales has claimed that it is entitled to the deduction under s 73B of the Income Tax Assessment Act 1936 (ITAA36) on the basis that it has incurred the requisite level of expenditure in respect of research and development activities carried on by it or on its behalf. The research and development activities conducted by RACV Sales in those years were generally described as:

"Development of Crash Testing Methodology, to be carried out as part of the Australian New Car Assessment Program (ANCAP) investigating the impact of [the] Automotive Structural Design and Safety Features on serious injuries and fatalities, and Associated Development of Use Car Safety Ratings (UCSR)."[7]

4. Innovation Australia decided on 4 October 2007 that RACV Sales' activities did not satisfy the eligibility criteria under s 73B(1) of ITAA36 as the activities did not involve innovation or high levels of technical risk. Therefore, it also decided to issue a certificate under s 39L of the IRD Act to reflect that decision. At the request of RACV Sales, Innovation Australia reviewed its earlier decision at its meeting held on 25 February 2010. Under s 39S of the IRD Act, it confirmed that earlier decision and commented that the activities were not systematic, investigative and experimental.[8] We have decided that Innovation Australia is correct in deciding that RACV Sales' activities are not research and development activities as those activities are defined in s 73B(1) of ITAA36. Therefore, we have decided to affirm its decision to issue a certificate under s 39L of the IRD Act to that effect.

5. For the purposes of this hearing and as providing part of the background to it but for no other purpose, we have assumed that RACV Sales forms part of the corporate structure associated with the Royal Automobile Club of Victoria (RACV). We have also assumed that RACV Sales and other corporate entities within that structure operate as part of a tax consolidated group.[9] The RACV, in one corporate guise or another, is a member of the Australian Automobile Association (AAA). Given that our role is limited to matters relevant to the decision made under s 39L of the IRD Act and not s 73B of ITAA36, we do not need to explore that issue further.

The activities for which RACV Sales is registered

6. The activities registered by RACV Sales may be categorised as crash testing activities conducted by RACV Sales through ANCAP; data interpretation and presentation activities conducted by RACV Sales through ANCAP; low speed crash testing activities conducted by IAG for RACV Sales and others; supporting activities conducted by RACV Sales through ANCAP; and supporting activities conducted by RACV Sales through the Australian National Crash In-depth Study (ANCIS).

7. The following table was provided by RACV Sales. It sets out the nature of the activity being conducted under the headings we have referred to above; the period during which those activities were conducted; and whether those activities are properly described as systematic, investigative and experimental (SIE) or directly related (DR):

Review of the existing international vehicle safety testing techniques and the applicability of these tests to Australian motor vehicles.

Assessment of the applicability of these techniques in Australia based upon consideration of the types of motor vehicle accidents that result in fatalities and serious injuries, and the nature of the Australian road and traffic environment.

Attendance, presentation and review of conference papers considering the issue of testing techniques utilised in global consumer crash test programs.

The first series of vehicle testing to utilise the full frontal test method only. Full frontal testing simulates the impact of a collision occurring evenly across the front of the vehicle. The objective of this test is to evaluate the vehicle's restraint system. The testing investigated the safety of a range of new, previously untested, vehicle models.

Expand test methods to incorporate a variety of different car types and styles - small car, compact 4WD, and large car. Review testing data as required for each test performed.

Continual development of data interpretation and presentation techniques

This activity involves the development of interpretations, which simulate the real world, and development of an appropriate format in which to convey the information obtained in these tests to consumers. It was determined that this information should contain both the testing results and the full technical report. In addition, a colour-coded human figure was included to indicate the extent of head and chest injuries.

Further investigation was undertaken to observe and research additional safety testing techniques, offset testing, being employed globally. The applicability of this testing technique in Australia was also investigated.

An offset test using a deformable barrier was added. The test is internationally recognised, and the barrier design is specified for both consumer crash test programs and for compliance with regulatory standards.

Further series of vehicle tests were undertaken, incorporating the offset test. Offset testing simulates a collision of one vehicle with another. The objective of this test is to evaluate the vehicle's structure resistance to intrusion; the crash forces are concentrated on the driver's side of the vehicle.

Further development of the National Crashed Vehicle Database

Establish a database of crashes in NSW, QLD and Victoria to look in detail at a small number of crashes each year.

Involves a pendulum device colliding with the front driver's side of a vehicle, equivalent to a barrier impact at 15 kph as a consistent method of rating the repairerability of a vehicle. By 1999 RACV Sales had provided test results for 27 models in use in Australia.

The test method is being introduced to all new vehicles, particularly to the application of small 4WDs and larger vehicles.

A trolley fitted with a deformable front is towed into the driver's side of the car to simulate a side on crash.

A specialised dummy is used in this test to collect maximum information on the scale of injury present. Readings taken from dummies are used to assess protection given to the front occupant.

It has been found in testing that this method is unsatisfactory for larger and 4WD vehicles, as the impact boom has a tendency to push underneath the sill panels, contact the chassis rails and in some cases tip the vehicle. This has returned problematic and inconclusive data. This has led to the further development of the side impact test-outlined in the development of the pole test.

Approximately a quarter of all serious-to-fatal injuries happen in side impact collisions. Many injuries occur when a car hits a pole or a tree.

To encourage manufacturers to fit head protection devices, pole/head protection tests have been added to ANCAP protocols. Side impact airbags help to make this kind of crash survivable. They are also very effective in other types of side impact accidents such as being hit by another vehicle where the bonnet enters the window at head height.

In the new test, the car tested is propelled sideways at 29 kph (18 mph) into a rigid pole. The pole is relatively narrow, so there is major penetration into the side of the car.

This development has been incorporated into the testing of larger vehicles and 4WDs. The hardware and test rig currently do not exist, RACV Sales will be developing the testing methods and equipment.

RACV Sales will be rolling out a three stage 4WD testing regime relating to various vehicles. This will enable validation of tests and results to provide meaningful crash performance information.

A series of tests are carried out to replicate accidents involving child and adult pedestrians where impacts occur at 40 kph (25 mph). Impact sites are being assessed and rated fair, weak and poor.

For 2004/2005, RACV Sales undertakes to further develop the pedestrian impact test which has to date been based on overseas programs. Further test development is required for local conditions, for example incorporation of pedestrian tests for higher vehicles and vehicles fitted with bull bars.

Continual development of key information presentations

The concentration on ANCAP injury content was considered to be too obscure for the intended audience and a simplified format was required to retain the integrity of that data. The following four information releases were amended to adopt a new primary rating system, that being the risk of life-threatening injury. The presentation of test results has evolved through a process of continuous review. After a review of the presentation methods adopted by a number of international organisations, a rating system was adopted which ranked the likelihood of serious injury as good, acceptable, marginal or poor. With the availability of additional information for consumers, further redevelopment was undertaken in relation to the manner in which the information was conveyed. It was determined that the most useful method for presenting this information would be through the calculation of an overall risk score, which would simulate real world frontal crashes.

Development and evaluation of interpretations with real world data

This involves the relationship between laboratory crash testing and real-world crashes in Australian conditions. A vehicle's crashworthiness is assessed on the basis of structure, restraints, injury measurements and head restraint design. This combined overall evaluation is determined on the basis of both the full frontal and offset crash test results.

Re-design and development of data interpretation and information presentation

Interpretations are evaluated based on a comparison with real-world data and interpretation and presentation techniques are redesigned.

Further investigations will be undertaken in relation to the possible addition of further testing which can assess vehicle safety in side impact, pedestrian impact, and vehicle rollover susceptibility.

Continual design of testing procedure and data manipulation

Incorporation of new testing techniques to assist vehicle safety. New tests performed assist in identifying vehicle safety in side impact and pedestrian impact. The pedestrian impact test was trialled for the first time in September 2000. The results from the tests were encouraging, and vehicle testers worldwide have mirrored the ANCAP research within their own vehicle evaluation procedures.

The small car program is the first stage of ANCAP testing and has been designed and developed to allow harmonisation with EuroNCAP testing procedures. The ANCAP harmonisation project saw ANCAP undertake a program whereby it reused test data from EuroNCAP tests and evaluated their applicability to European cars available for sale in Australia. The technical committee learnt many valuable lessons from the first and subsequent launches of this program, for example, crash tests cannot be directly transferred from European conditions to Australian conditions. In many cases standards, components and environmental conditions can vastly influence the test result. Modifications need to be made to the tests and the analysis to allow for factors such as differences in passenger restraint systems between Australian and European models.

Pedestrian impact tests were included for the first time. These tests have subsequently become an adopted standard with many of the NCAP Programs worldwide.

Large cars account for approximately one third of all new car sales in Australia. It is a rapidly changing market with new models and safety systems being released at an alarming rate. For ANCAP to be of any use it must test and analyse these new systems in a timely and efficient manner.

2000/2001 has seen significant occupant protection upgrades across all vehicles within the large car market. These upgrades must be assessed by ANCAP and analysed. Often the testing and analysis equipment must be reconfigured to adapt to new testing parameters arising from new vehicle components.

A new testing procedure has been designed for utilities. Utilities have not been assessed since 1995, and the market has increased significantly in this time. This series will also be used to investigate the effect of sill height on the response of vehicles in a side impact test, e.g. 4WD utility vehicles.

Vehicles with H-points above or below the 700 mm threshold specified in ADR 72 (Australian Design Rules) for exemption will be tested and information from this will be used in the development of an approach to testing other 'high vehicles' such as 4WDs.

RACV Sales will undertake development of a new locally developed roll over test. The test will be dynamic, in that it is based on a ratio of centre of gravity and track width.

US tests are based on a 'fish hook' methodology - the vehicle is driven at speed around a fish-hook shaped test track with increasing velocity on a test bed until it is on the verge of rollover.

With the onset of mainstream vehicle electronic stability control systems, RACV Sales has identified the need to develop testing regimes that can assess ESC systems for new vehicles. Currently the development is at infancy, further basis for the tests will follow in the 2004/05 year and beyond.

Review of divergence from European Testing Procedures

Development of new processes to overcome the divergence of Australian vehicle testing requirement methods from European and offshore testing. In particular new testing processes, the shortfalls of existing processes will need to be developed to overcome the specific parameters of the Australian market, including the prevalence of 4WD vehicles and the fact that Australian vehicles are not as well specified as European models.

8. As we have said, RACV Sales did not carry on any of the activities itself. For the purposes only of setting the background to the limited issues we must decide, we have made findings relating to the way in which ANCAP, ANCIS and IAG operated and RACV Sales' involvement with them.

9. On the basis of the evidence of Mr Michael Case,[10] we find that ANCAP was established on 30 May1992 by the Roads and Traffic Authority of New South Wales (RTA), Queensland's Department of Transport (Qld DOT), the Roads Corporation of Victoria (VicRoads), South Australia's Commissioner for Highways (SADRT) and the Australian Automobile Association (AAA). Each contributed funds of various amounts:

"... for the operation, management and development of a program for crash testing, collating and publishing the relevant performance of the total occupant protection system of popular motor vehicles sold on the Australian market with the aim of facilitating improvement in motor vehicle occupant protection through consumer education and buying power influence."[11]

10. Since then, the membership of the consortium has expanded to include all State and New Zealand Transport Departments, all Australian automobile clubs through the AAA, the New Zealand Automobile Association, the Victorian Transport Accident Commission and the FIA Foundation.[12] Between 1999 and 2006, the RACV first contributed to ANCAP by means of funds and participation through the AAA and later as a Contributing Member to ANCAP. It also contributed services in the form of work undertaken for ANCAP by Mr Case.[13]

11. In 2004, Monash University, the AAA and the RACV together with other State transport authorities, the automotive industry and insurance and consumer groups formed a consortium to finance and support "The Australian National Crash In-depth Study" (ANCIS). The study was to be undertaken by the Accident Research Centre at Monash University. Its purpose was described at cll B and C of the Consortium Agreement:

"B. ANCIS was formed to provide detailed in-depth data on a random sample of vehicle crashes that occur in a number of states in Australia ('the Data'). Its objectives are to undertake a retrospective in-depth examination of a sample of passenger vehicle crashes where at least one occupant was injured sufficiently to be hospitalised to determine the extent of the damage to the vehicle, the severity of the crash, the level of injuries and the cause of the injuries to the occupants and to undertake a retrospective examination of the likely factors associated with the crash.

1. The underlying purpose of the collection of data is for research and the data will be used primarily for improvement of vehicle occupant safety through vehicle regulation, design, infrastructure and road user safety initiatives ..."[14]

12. IAG is the Insurance Australia Group. It was previously known as Insurance Australia Group Limited. Before that, it had been known as the NRMA Insurance Group Limited, which had been formed when the National Roads and Motorists Association (NRMA) demutualised in 2000.

13. Since RACV Sales first applied for registration in respect of the 1988-89 income years, the IRD Act and ITAA36 have been amended on several occasions. Those amendments may be relevant to the assessment of any deduction to which RACV Sales may be entitled but we have referred only to amendments to provisions that are either relevant to the issues we must decide or set the background against which they are decided. We have done so at Attachment A to these reasons and it forms part of these reasons no less than any of its other paragraphs. It sets out the context of the relevant provisions, which informs an understanding of them but, at this point, we have set out only the pivotal provisions.

(1) An eligible company may claim a deduction from its assessable income when it incurs "research and development expenditure": ITAA36, s 73B(14).

(2) "Research and development expenditure" includes expenditure incurred directly in respect of "research and development activities carried on by or on behalf of the company": ITAA36, s 73B(1).

"(a) systematic, investigative and experimental activities that involve innovation or high levels of technical risk and are carried on for the purpose of:

(i) acquiring new knowledge (whether or not that knowledge will have a specific practical application); or

(ii) creating new or improved materials, products, devices, processes or services; or

(b) other activities that are carried on for a purpose directly related to the carrying on of activities of the kind referred to in paragraph (a).": ITAA36, s 73B(1).

(4) "... the following activities are taken not to be systematic, investigative and experimental activities:

(a) market research, market testing or market development, or sales promotion (including consumer surveys);

(d) the making of cosmetic modifications or stylistic changes to products, processes or production methods;

(i) routine collection of information, except as part of the research and development process;

(n) any activity related to the reproduction of a commercial product or process by a physical examination of an existing system or from plans, blueprints, detailed specifications or publicly available information.": ITAA36, s 73B(2C).

(5) "For the purposes of the definition of research and development activities in subsection (1):

(a) activities are not taken to involve innovation unless they involve an appreciable element of novelty; and

(b) activities are not taken to involve high levels of technical risk unless:

(i) the probability of obtaining the technical or scientific outcome of the activities cannot be known or determined in advance on the basis of current knowledge or experience; and

(ii) the uncertainty of obtaining the outcome can be removed only through a program of systematic, investigative and experimental activities in which a scientific method has been applied, in a systematic progression of work (based on principles of physical, biological, chemical, medical, engineering or computer sciences) from hypothesis to experiment, observation and evaluation, followed by logical conclusions.": ITAA36, s 73B(2B).

(6) "If the Board gives to the Commissioner a certificate stating whether particular activities carried on by or on behalf of a specified eligible company were research and development activities, that certificate is binding on the Commissioner for the purpose of making an assessment of the company's taxable income of any year of income in which those activities were carried on.": ITAA36, s 73B(34) complementing s 39L of the IRD Act.

(1) Part IIIA of the IRD Act complements s 73B of ITAA36 by giving Innovation Australia the role of determining whether an eligible company satisfies the requirements for the incentive in the form of the deduction available under that provision: IRD Act, s 39AA(1).

(2) "(1) The Board may, and shall if requested in writing by the Commissioner to do so, give to the Commissioner a certificate stating whether particular activities that have been or are being carried on by or on behalf of a person were or are research and development activities.

(2) If the Board issues a certificate to the effect that particular activities were not or are not research and development activities, the Board must give notice in writing to the person concerned stating the reasons for issuing the certificate.": IRD Act, s 39L.

16. On 25 February 2010, the Board confirmed its decision dated 27 September 2007 that:

"The activities undertaken in the project entitled 'RSM1 - Understanding the Results of Crash Testing and the Impact of Automotive Structural Design and Safety' carried out in the 1998-99 to 2005-06 years of income do not satisfy the eligibility criteria under section 73B(1) of the Income Tax Assessment Act 1936, as the activities did not contain innovation or high levels of technical risk."[15]

Scope of decision under review: one project or two

17. RACV Sales applied for registration in relation to activities described as the "RSM1 project" in respect of the years 1998-99 to 2003-04 but, in the last two years, was registered in relation to very similar activities under the project name of "RACV05-01". RACV Sales and their legal representatives have treated them as the one and the same project and their activities as one and the same. The Board has recognised that the activities were described in materially identical terms[16] but its decision gives no recognition to RACV05-01.

18. As s 39T(1)(a) provides that an application may be made to the Tribunal for review of a decision that has been confirmed under s 39S(4), that means that the application is made, in this case, in respect of the Board's initial decision made on 27 September 2007. In view of the Board's reference to all of the years and on being satisfied that the activities under each project were described in materially identical terms, we have understood the decision under review to refer to the activities whether described under one project name or the other.[17] This view is consistent with the scheme of registration provided for under s 39J of the IRD Act. It allows registration only "in relation to those research and development activities, in respect of that year of income"[18] and so only those for which registration has been sought. In the 2004-05 and 2005-06 years, RACV Sales sought registration only for the activities described under the RACV05-01 project and not under the RSM1 project. As the Board could have registered only the activities for which registration was sought, its certificate under s 39L could relate only to those activities and so only to RACV05-01 activities in those two years. It would seem, therefore, that the reference to the activities under RSM1 in the decision should be read as a reference to the activities under the relevant projects in relation to which registration was sought in each of the years.

Scope of the decision under review: meeting the eligibility criteria under s 73B(1) of ITAA36 or under the IRD Act?

19. The Board's decision refers to the relevant activities' not satisfying "the eligibility criteria under section 73B(1)". We are a little uncomfortable with framing that issue in this way because it tends to suggest that the decision is being made under ITAA36 when it is acknowledged that it has been made under s 39L of the IRD Act.

20. The issue under s 39L(1) is whether the "particular activities that have been or are being carried on by or on behalf of a person were or are research and development activities." In view of ss 39A(2) and (3), what amount to "research and development activities" is defined by reference to s 73B of ITAA 36 unless the contrary intention appears from Part IIIA of the IRD Act. Other expressions are defined by reference to their definitions in provisions of ITAA36 other than s 73B.

21. Certainly it is the case that, since 2001, s 39AA(1) has stated that the object of Part IIIA, and so of the Board's role under s 39L, is to complement the tax incentive provided under, among others, s 73B of ITAA36. Having a role to complement a tax incentive does not mean that the Board makes a decision under the provision under which that tax incentive is provided. It does not mean that it is making a decision applying the criteria specified in ITAA36. Its decision must be made within the confines of the IRD Act and so within the confines of its definition of "research and development activities". Had the IRD Act provided that the Board was to make its decision under s 73B(1) of ITAA36, it would have been obliged to do so. There is no such provision. All that s 39A(2) has done is to ensure that, in interpreting what amounts to "research and development activities" under the IRD Act, the Board does so by reference to the meaning of the expression set out in s 73B. That does not mean that the Board makes its decision under s 73B or that it is deciding whether it is satisfied whether an eligible company meets "the eligibility criteria under section 73B(1)". It is deciding whether the particular activities carried on or being carried on by an eligible company are research and development activities. It is making that decision under s 39L by reference to the provisions of the IRD Act which are to be interpreted for some purposes by reference to s 73B.

22. Having regard not only to the definition of the expression "research and development activities" in s 73B(1) but also to the qualifications to it in ss 73B(2B), (2BA) and (2C), the following steps need to be taken in deciding whether or not activities are research and development activities:

(1) Have the activities been, or are they being, carried on by or behalf of a person[19] as required by s 39L of the IRD Act?

(a) If they are not - as would be the case if they were merely proposals - the Board has no basis on which to issue a certificate under s 39L of the IRD Act; and

(b) if they are, the Board does have a basis on which to issue that certificate.

Paragraph (a) of the definition of "research and development activities" in s73B(1)

(2) Are the activities properly characterised as "systematic, investigative and experimental activities ..."?

(a) If the activities are activities specified in s 73B(2C), they are not "systematic, investigative and experimental activities ..."; and

(b) if they are not specified in s 73B(2C), they must be characterised.

(3) If the activities are not properly characterised as "systematic, investigative and experimental activities ...", they are not research and development activities within the scope of paragraph (a) of the definition of "research and development activities" in s 73B(1).

(4) If the activities are properly characterised as "systematic, investigative and experimental activities ...", do they "... involve innovation or high levels of technical risk"?

(a) Whether activities involve innovation must be decided by reference to s 73B(2B)(a); and

(b) whether activities involve high levels of technical risk must be decided by reference to both limbs of s 73B(2B)(b) and having regard to s 73B(2C).

(5) If the activities are not properly characterised as involving innovation or high levels of technical risk, they are not research and development activities within the scope of paragraph (a) of the definition of "research and development activities" in s 73B(1).

(6) If the activities are properly characterised in that way, are they carried on for the purpose of either:

(a) acquiring new knowledge (whether or not that knowledge will have a specific practical application); or

(b) creating new or improved materials, products, devices, processes or services?

(7) If the activities are carried on for neither purpose, they are not research and development activities within the scope of paragraph (a) of the definition of "research and development activities" in s 73B(1).

Paragraph (b) of the definition of "research and development activities"

(8) If they do not come within the scope of paragraph (a) of the definition of "research and development activities" in s 73B(1), the question becomes:

(a) Are they activities that are carried on for a purpose directly related to the carrying on of activities of the kind referred to in paragraph (a) of the definition of "research and development activities" in s 73B(1)?

(i) If they are, they are research and development activities within the scope of paragraph (b) of the definition of "research and development activities" in s 73B(1).

(ii) If they are not, they are not research and development activities within the scope of paragraph (b) of the definition of "research and development activities" in s 73B(1).

Exclusion of activities otherwise coming within definition of "research and development activities"

(9) Are the activities on and after 1 July 2002 excluded by s 73B(2BA) on the basis that they were not carried on in accordance with a plan complying with guidelines formulated by the Board under s 39KA of the IRD Act?

(a) This criterion applies only to the years from 2002-03 to 2005-06 and not the earlier years.

(a) were not so carried on, they are not research and development activities within the definition of "research and development activities" in s 73B(1); or

(b) were so carried on, they are research and development activities within the definition of "research and development activities" in s 73B(1).

23. Our role is confined to a consideration of whether the particular activities in relation to which RACV Sales has sought registration or been registered and which we have identified at [7] above have been, or are being, carried on by RACV Australia and, if so, whether they are research and development activities. It is not to decide whether RACV Sales is entitled to a deduction under s 73B of ITAA36. That is a matter for a separate decision by the Commissioner and, if an application is made to it, of separate review by the Tribunal.

SUMMARY OF CONCLUSIONS OF ANALYSIS OF THE LEGISLATIVE PROVISIONS AND CASE LAW

24. At Attachment B, we have set out a detailed analysis of the provisions and the case law. Again, that Attachment forms part of our reasons and underpins our consideration of the evidence in the following section of our reasons. In this section, we set out some of our main conclusions:

(1) Criteria identifying "research and development activities"

• "systematic, investigative and experimental activities"; and
• "involve innovation or high levels of technical risk"; and
• are carried on for the purpose of acquiring new knowledge or creating new or improved materials, products, devices, processes or services;

• "other activities carried on for a purpose directly related to the carrying on of activities of the kind referred to in paragraph (a)."

Regard must be had to s 73B(2B) in determining whether the activities involve innovation or high levels of technical risk.

(b) In order to understand their import, the background in which they were carried out must be understood.

(c) We are concerned with characterisation of the activities and not with that of the project of which they form a part.

(3) When are activities "systematic, investigative and experimental"?

(a) The use of the conjunction "and" between the second and third signifies that each of the words qualifies the activities i.e. they must be "systematic, investigative and experimental activities" (emphasis added). It is not enough, for example, to be systematic and investigative if the activities are not also experimental. They must meet all three descriptors and one descriptor does not qualify another.

(b) The qualifications set out in s 73(2C) to the activities that are regarded as "research and development activities" limit the activities but do not limit the interpretation of those words in the first instance.

(c) The words "systematic, investigative and experimental activities" should be given their ordinary meanings:

"systematic": "... 1 making use of, or carried out according to, a clearly worked-out plan or method. 2 methodical ..."[20]

"investigative":"investigate ... verb ... to carry out a thorough, detailed, and often official inquiry into, or examination of, something or someone. ... investigative ... adj. ..."[21]

"experimental""... 1 consisting of or like an experiment. 2 relating to, or used in, experiments. 3 trying out new styles and techniques. ..."[22]

["experiment" "... 1 trial carried out in order to test a theory, a machine's performance, etc or to discover something unknown. 2 the carrying out of such trials. 3 an attempt at something original. ..."[23]]

(4) When do systematic, investigative and experimental activities "involve" innovation or high levels of technical risk?

(a) In its context, the word "involve" means "to require as a necessary part. ... " rather than simply "To include, contain or comprehend within itself or its scope" to that of "to require as a necessary part. ... ".

(b) That does not mean that every part of the activity must be a necessary part of, and so involved in, innovation or high levels of technical risk but, when viewed as a whole, the particular activities under consideration must have as a necessary part of them, and so involve, innovation or high levels of technical risk.

(5) When do systematic, investigative and experimental activities involve "innovation or high levels of technical risk"?

(a) The word "innovation" as it appears in paragraph (a) of the definition of "research and development activities" in s 73B(1) of ITAA36, should be given its ordinary meaning i.e. something new be it a new method or process, a new thing, or new knowledge. It must have that element of "newness" if it is to be innovative.

(i) We do not agree with the statement at [9.55] in the Explanatory Memorandum that it is relevant to consider whether "the activities are likely to result in patentable or other protectable intellectual property."

(b) Cases either decided before or dealing with the law before its amendment on 23 July 1996[24] are not relevant in deciding whether the ordinary meaning of "innovation" is qualified by the addition of s 73B(2B)(a) on that date i.e. by the requirement that "activities are not taken to involve innovation unless they involve an appreciable element of novelty".

(c) The expression "an appreciable element of novelty" should not be interpreted by reference to any similar concept in patents law.

(d) It should be interpreted according to its ordinary meaning and so require the following questions to be asked and answered after viewing the activities as a whole and in context:

(i) Do the activities claimed to be research and development activities have a feature(s) or compent(s) that is new (and so an element of novelty);

(ii) If so, is that element of novelty something that is significant in some way (appreciable)?

(6) Activities must involve high levels of technical risk

(a) The ordinary meaning of the expression "technical risk" is that there is a chance or possibility that the processes or means (be they classified as mechanical, electrical, scientific, technological or in some other way that can be described as "technical") involved in the activities will fail to work, to play the part intended for them or to achieve an outcome intended for them.

(b) How failure is described and what amounts to failure will depend upon the particular technical discipline concerned. Whether there is a chance or possibility of failure depends upon evidence relevant to that discipline.

(i) Technical processes or means may themselves fail and yet, for reasons not necessarily foreseen, achieve their purpose. Achievement of the purpose for which the activities are carried out is not a relevant consideration in assessing technical risk.

(c) In order to establish "high levels of technical risk" there must be a significantly greater chance or possibility that the technical processes or means involved in the activities will fail than would normally be the case.

(i) What would normally be the case will be ascertained by reference to the industry or field of endeavour in which the activities are being conducted.

(iii) What will normally be the case, and so the level of risk, cannot be measured by reference to legal standards such as whether it is "more likely than not" that there will be technical failure.

(d) In determining the qualifications in s 73B(2B)(b):

(i) Whether the probability of obtaining the technical or scientific outcome of the activities cannot be known in advance on the basis of current knowledge or experience is an objective test i.e. it is not determined by reference to the knowledge or experience of those undertaking the activities.

(ii) Whether the uncertainty of obtaining that technical or scientific outcome can be removed only through a program as described in s 73B(2B)(ii) requires:

• a "program" and so a plan or schedule of "systematic, investigative and experimental activities";
• "scientific method" must be applied in that program and so in a manner according to "an ordered set of procedures or an orderly system. ..."[25] as it would be in science and "... displaying the kind of principled approach characteristic of science". The reference to science does not limit the method to any particular activity for it extends to "... any area of knowledge obtained using, or arranged according to, formal principles. ..." but must be based on principles of physical, biological, chemical, medical, engineering or computer sciences.

(iii) The program must be conducted in an orderly way beginning with an hypothesis i.e. "... 1 a statement or proposition assumed to be true for the sake of argument. 2 a statement or theory to be proved or disproved by reference to facts. 3 a provisional explanation of anything."[26]

(iv) It is implicit that the hypothesis be framed so that it is relevant in, or directed to, removing the uncertainty of obtaining the technical or scientific outcome of the activities. That means that it is relevant in, or directed to, determining the probability of obtaining the technical or scientific outcome of the activities.

(v) There must be a "progression" of that work so that it can be said to be "... moving forwards or advancing in stages ..." to "experiment, observation and evaluation".

• the experiment, observation and evaluation must be directed to testing the hypothesis;
• what work amounts to experiment, observation and evaluation depends on the particular discipline in which it is being undertaken and is a matter for evidence.
• having done so, the work must move to drawing logical conclusions from the experimental, observational and evaluative work. Those logical conclusions must relate back to the hypothesis.

(7) "activities ... carried on for the purpose of" acquiring new knowledge or improved materials etc"

(a) The purpose for which activities are carried on is the object or aim for which they were carried on at the time.

(i) It is not determined according to a rationale developed at a later time to explain why those activities were carried on.

(b) We do not need to decide in this case whether the purpose need be the sole purpose, a dominant purpose etc.

(c) The activities must be carried on for the purpose of:

(i) acquiring new knowledge in the sense of materials not previously known or discovered or creating materials not previously known or, in the case of materials, materials of a higher or better quality or value.

(8) "other activities that are carried on for a purpose directly related to the carrying on of activities of the kind referred to in paragraph (a)"of the definition of "research and development activities"

(a) The requirement that the activities be "of the kind referred to in paragraph (a)" must be a reference to the activities being "of the same sort" as those referred to in paragraph (a) of the definition of "research and development activities".

(i) A taxpayer may seek a deduction in respect of other activities referred to in paragraph (b) of the definition and not of those referred to in paragraph (a). That might arise if, for example, the taxpayer were not carrying on the activities of the kind referred to in paragraph (a).

(b) The purpose for which the other activities are carried on need not be the sole, dominating or actuating purpose.

(c) Whether and activity is carried on for a purpose directly related to the carrying on of activities in paragraph (a) of the definition of "research and development activities":

(ii) it will be relevant to consider whether there are other factors that need come into play or steps that need to be taken before it can be said that one activity is carried on for a purpose related to another.

25. In light of our analysis of the legislative provisions and case law, we have examined the activities which RACV Sales has described in the table at [7] above more carefully in order to determine the precise nature of each activity. We will commence with the SIE (systematic, investigative and experimental) activities as the DR (directly related) activities must be related to those SIE activities we find satisfy the legislative requirements for research and development activities.

Activity 3: Continual development of data interpretation and presentation techniques

26. While the heading describing the research and development activities refers only to the development of data interpretation and presentation techniques which simulate the real-world, the activity to which RACV Sales referred appears to also involve investigation undertaken to observe and research additional safety testing techniques. The additional safety testing techniques include using a deformable barrier incorporating an offset test, which simulates the collision of one vehicle with another, the impact being recorded on the driver's side of the front of the vehicle.

27. Mr Michael Colin Case, who is the Manager, Vehicle Engineering at RACV Sales, explained in his supplementary witness statement made on 16 August 2011 that what is required of ANCAP is to ensure its safety ratings correlate with real-world crashworthiness for given motor vehicles. He said that the following questions needed to be posed:

"(a) how well the test protocols and configurations simulate real-life crash scenarios, having regard to variables like common accident types in Australia and the Australian vehicle mix;

(b) do the test protocols and configurations produce undesirable results, such as prioritising structural integrity at the expense of aggressivity to other vehicles and pedestrians;

(c) is sufficient data recorded or should additional information be taken into account in assessing crash test performance, including by way of modification; and

(d) are there safety devices and features that the ANCAP rating should specifically give weight to that because they will reduce the risk of injury and fatalities in Australian road traffic accidents?"

28. Mr Case said that between 1999 and 2006 RACV Sales, through ANCAP, investigated all of these issues to establish whether there were compelling safety and technical reasons to either deviate from the harmonised testing protocols or to suggest improvements to those protocols. Harmonised testing protocols were introduced in 1999 so that ANCAP could use EuroNCAP ratings for imported vehicles which had identical specifications to the vehicles already tested by EuroNCAP. This would avoid duplication of effort and additional costs. In a June 1998 article entitled International NCAP Programs in Review, which was co-authored by Mr Case, the authors state that ANCAP recognised the global nature of the car manufacturing industry and developed a crash testing program to align with the existing overseas programs. The article states that in addition to a full frontal test, an offset test using a deformable barrier was added in 1994. This test was developed for the European Experimental Vehicle Committee (the EEVC) by the Transport Research Laboratory in the United Kingdom.

29. The article co-authored by Mr Case also refers to the presentation of results which has evolved by a process of continuous review, including focus group evaluations. At that time (1998), the presentation format used ratings of good, acceptable, marginal and poor. That format was considered to be the best available by organisations visited during an international ANCAP study tour.

30. In addition, the article refers to four areas of vehicle crashworthiness which were evaluated. They were vehicle structure, restraints, injury measurements and head restraint design. The authors also noted that the ANCAP rating based on data provided by overseas organisations might differ from the ratings assigned by these organisations. That appeared to be because the ANCAP rating included assessment of the results of full frontal crash tests and took into account passenger injury measures and restraint performance in these tests. ANCAP ratings tended to place less emphasis on footwell intrusion and lower leg injury and more emphasis on structural performance. The authors recognised the desirability of having the same rating system used by all organisations but stated that this would have several disadvantages for the Australian new car fleet.

31. In a witness statement dated 15 August 2011 Mr Michael Philip Paine, a mechanical engineer who has provided consulting services to ANCAP since 1996, said that ANCAP uses essentially the same test protocols as EuroNCAP and has sought to do so since harmonisation in 1999. Attached to Mr Paine's witness statement was a copy of ANCAP notes on the assessment protocol as at April 2011. Those notes state that ANCAP conducts crash tests and associated assessments in accordance with the test protocols issued by EuroNCAP. Mr Paine noted that ANCAP ceased conducting a full frontal crash test following harmonisation with EuroNCAP in 1999.

32. The protocols for crash testing and the scores allocated to individual tests may vary from the scores allocated by EuroNCAP. A star rating of five stars maximum is used by both ANCAP and EuroNCAP. As is set out in the ANCAP Notes issued in April 2011, where a star rating is limited by an individual score, ANCAP reduces the overall score to the maximum that is available for that star rating. However, for a vehicle to be awarded a five-star rating, it must meet the additional ANCAP requirements, being the earning of at least one point in a pole impact test and vehicles must be equipped with an electronic stability control (ESC) system that meets the requirements set out in the Notes. ANCAP also has a departure from EuroNCAP testing and assessment protocols for side impact protection for occupants of high seat vehicles as those vehicles are common in Australia but not in Europe. For all vehicles, including high seat vehicles, a successful pole test is required to achieve five stars. Where ANCAP uses crash test data from EuroNCAP, the pole test score will only be included in the overall score where a vehicle has achieved at least a four-star rating without a pole test.

33. In his supplementary witness statement Mr Case said that each vehicle tested by ANCAP follows the EuroNCAP test protocols for the configuration, speed and other aspects of the conduct of each test. Data such as acceleration, force and displacement are measured through sensors in various locations on a fully instrumented dummy (the EuroNCAP Euro SID II) which is captured by on-board data acquisition equipment. In his witness statement Mr Paine explained that data from the data acquisition unit is downloaded to a computer which has algorithms to process it and produce injury values such as Head Injury Criteria (HIC). The scoring and rating process assigns a score to the injury values for each body region. Further analysis is conducted of video information and in the case of the offset test, measurements of intrusion of the steering wheel, pedals and the cabin structure are taken into account. These may result in deductions from the injury scores. The scores are evaluated to give an overall star rating.

34. Mr Paine said that after considering all the data and measurements, as well as making a record of his own observations from the crash tests, he analysed the data in accordance with the Assessment Protocol and prepared a scoresheet with a provisional overall score and star rating. This scoresheet is then provided to the manufacturer of the vehicle and a public datasheet is prepared which becomes the ANCAP webpage for that vehicle. Mr Paine said that where he makes observations concerning a vehicle's performance in a crash test that is outside the assessment protocol, he includes those observations in a detailed report provided to the car manufacturer. He gave the example where a fuel tank was damaged by the tail shaft in a crash test. He said that from time to time ANCAP reported such safety-related equipment failures to the relevant federal department for possible further investigation. He said that several vehicle recalls have resulted from this process.

35. Mr Paine also explained that ANCAP has sought to improve the format and presentation of its public datasheet to convey information to consumers as effectively as possible. Attached to his witness statement were a series of reports on different vehicles demonstrating the change in the format used for earlier public datasheets. The significant difference between the various datasheets is that the earlier publications of test results (1998-1999) give ratings indicating good, acceptable, marginal or poor while from 2000, the test data provides a star rating. Furthermore, the number of injury measurements taken has increased to include injuries to the neck, abdomen and pelvis.

36. In his witness statement Mr Paine said that the tests undertaken by ANCAP seek to simulate only four crash scenarios. Since the late 1980s, the Monash University Accident Research Centre (MUARC) has analysed crash statistics (real accidents) and produced estimates of risk of serious injury to drivers of most models of car in Australia. These are known as Used Car Safety Ratings (UCSR). Soon after ANCAP commenced crash tests, MUARC began analysing links between USCR and ANCAP ratings. Mr Paine said that he conducted informal analyses of USCR and the individual components of the ANCAP scoring system to check that an appropriate balance was given to the scoring process.

37. Mr Paine explained that the problem with ANCAP testing is that it seeks to simulate only four crash scenarios. In contrast, real-world vehicle crashes can take endless permutations, involving different vehicle numbers and types, impact areas, causes and consequences, such as rollover. However, subject to those limitations, Mr Paine said he compared the results of ANCAP tests with the outcome of real-world crashes and noted that the comparisons showed a reasonable correlation between ANCAP's crash testing and assessment results and real-world crash data. However, Mr Paine admitted there was a limit to the use to which ANCAP can put the information produced by MUARC, as the real-world crash data is, by its nature, retrospective, with a typical delay of about three years before the new model has had sufficient number of serious crashes to be statistically valid under the UCSR methodology. Furthermore, he said that ANCAP does not necessarily wait for real-world crash data to become available before setting its policy.

38. Innovation Australia contended that RACV Sales had not provided any contemporaneous records, including records of any experiments, which would enable the Tribunal to be satisfied that Activity 3 involved systematic, investigative and experimental activities. It contended that RACV Sales' description of this particular activity was at such a high level of generality that it was not possible to identify what was actually done. That is to say, it was not possible to identify the activities undertaken. It referred to phrases used by RACV Sales such as development of interpretations which simulate the real world and further investigations... to observe and research additional safety techniques, offset testing, employed globally.

39. Professor Manfred Zockel, a highly qualified mechanical engineer with some 40 years' experience in the areas of research, academia and industry, provided a witness statement dated 28 September 2011 which was taken into evidence. Professor Zockel said that in the course of his research career, he had undertaken a range of projects in which he proposed hypotheses, designed experiments to test those hypotheses and undertook those experiments. He retired as an Associate Professor and has held the title of Honorary Research Fellow in Mechanical Engineering at the University of Adelaide. In our opinion, Professor Zockel is well qualified to offer an opinion about experimental research.

40. Although Professor Zockel did not specifically address Activity 3, he made some general comments about the development of existing processes or protocols. Professor Zockel said that in his opinion, to demonstrate substantial development of existing processes or protocols, there should be test data from the application of a range of variables to show how the new or different test specifications and protocols have been developed. He said that in the material provided to him, he could not find any test data from which new or different test specifications and protocols were developed.

41. In the course of his cross-examination by Mr G McCarthy of counsel, who appeared on behalf of Innovation Australia, Mr Paine agreed there were three parts to the work conducted by ANCAP: the conducting of the tests; the recording, analysing and assessing of that data; and the presentation of the data to consumers. Mr Paine said that was the routine part of ANCAP's work. Mr Paine also agreed that the breaking down of the 14 activities was something which RACV Sales had done rather than the way in which those activities were viewed by ANCAP.

42. Mr McCarthy referred Mr Paine to an article which he co-authored entitled Crash Testing for Safety - Possible Enhancements to ANCAP Test and Rating Methods. The article, which is undated, appears to have been published in November 2008. In it the authors stated it became evident that the side impact test was not particularly demanding and most vehicles scored well in that test. It was also observed that this effect was, in some cases, disguising poor frontal offset test results. The authors stated that ANCAP had raised these concerns with EuroNCAP and suggested that a minimum score should be required for each type of crash test in order to earn a certain star rating, in addition to overall score criteria. Mr Paine agreed that this was a policy decision about how ANCAP would go about assigning certain star ratings according to the number of points that had been given, and to limit the ability of a vehicle to get a higher star rating than ought to have been given by doing well in the side impact test which was not worth much in any event. According to Mr Paine it was a policy decision to control the weighting given to scores achieved on the two tests. The article co-authored by Mr Paine also stated that ANCAP considered increasing the points for a good pole test to encourage manufacturers to fit head protecting airbags but ultimately decided to remain aligned with EuroNCAP.

43. Mr Paine also agreed that as part of the harmonisation with EuroNCAP, ANCAP adopted the same pedestrian protocols as those used by EuroNCAP. He also agreed that while ANCAP had a slightly different system for giving star ratings for the pedestrian test, it was nevertheless in accordance with the memorandum of understanding which gave rise to harmonisation.

44. In re-examination Mr Paine was asked to describe the non-routine part of ANCAP's work. He described the process as identifying vehicles which should be crash tested, the arranging of the crash tests, acquiring vehicles and organising the test laboratories. He said that after the tests were conducted and ANCAP received injury data and reports produced by the test laboratories regarding intrusion and any other observations about structural issues, an employee would look for what is described in the protocol as modifiers which would then be used to deduct points from the scores to reflect injury hazards which were not picked up in the dummy measurements. Mr Paine was then asked whether he would describe the information in the knowledge that has come from that as new he said: that's - those steps are all, as I say, fairly routine and straightforward. I've developed forms to try and make them as objective and quantitative as possible.

45. In cross-examination Mr Case agreed with Mr McCarthy's suggestion that the giving of star ratings, the scoring and the data collection techniques are consistent with those used by EuroNCAP. Any deviations from what Euro NCAP does is reflected in the protocol.

46. While we have no doubt that the way data collected from crash testing is interpreted and presented is significant in producing useful results from crash testing, it appears to us that the process is deductive rather than innovative. It is a process of determining from the data how to best present the information so that it accurately reflects the risk of injury to the occupants of a motor vehicle involved in a crash or a pedestrian who might be struck by a motor vehicle. Although the presentation of the data between 1998 and 2006 has varied in content and in form, we are unable to detect evidence of innovation. There is no new method or process, a new thing or new knowledge.

47. Essentially, points are awarded according to the values of a number of measurements taken both from the dummy representing a vehicle occupant, and the deformation of the vehicle; the points being weighted to reflect the risk of serious injury and a star rating assigned to the weighted points. There was no evidence of any test data or testing of the process of data interpretation or presentation techniques as one might expect if the process were innovative.

48. In our opinion, Mr Paine expressed it accurately when he referred to the process of recording, analysing and assessing the data and the presentation of the data to consumers as routine. Furthermore, since the adoption of EuroNCAP Protocols following harmonisation, data interpretation and presentation appears to have been largely standardised with ANCAP following the established protocols. We find that this activity does not satisfy the definition of research and development activities as defined in s 73B(1) of ITAA36 because it does not involve innovation or high levels of technical risk. It does not involve something new or new knowledge. Nor does it involve an appreciable element of novelty. There is no technical risk as the activity involving data interpretation and presentation must, even if not an accurate account of real-life accident outcomes, necessarily produce the outcome intended.

49. On its application form for registration for the income years 2000, 2001 and 2002, RACV Sales stated it was involved in the development of low speed crash testing. RACV Sales stated that crash test procedures involved a unique weighted pendulum device colliding with the front of the car. The weight impacted with the front driver's side, resulting in a crash equivalent to a barrier impact at 15 km/h.

50. In his witness statement dated 27 June 2011 Mr Case stated that this test was developed by IAG on behalf of RACV Sales and others. He said that the low speed test was not simply a test for safety but also for assessing the reparability of vehicles and as a consequence had an impact for consumers on costs and other factors such as insurance premiums.

51. According to Mr Case, the unique pendulum suspension system was developed by IAG. The system suspends the impact barrier, which is mounted on a large bucket, from four suspension points in a parallelogram configuration. This enables the impact barrier to be accurately swung in front of the test vehicle. He said that this system differs from those used overseas.

52. In his witness statement dated 23 September 2011, Mr Anthony Hyde, a former Director of Engineering and Design with GM Holden, referred to the pendulum test and said it was a common vehicle crash test which was in use in Australia and other parts of the world well prior to the early 1990s. He said the pendulum was used to simulate impact forces generated by a vehicle involved in a low speed collision. Vehicle manufacturers used the pendulum system to test bumper systems and, more recently, to verify airbag deployment calibrations because it could be accurately repeated and controlled in terms of speed, size and point of impact. One of the main users of the pendulum test is the insurance industry because insurers want to know results of low speed crash testing to assess the costs of repair of low impact crashes in order to set their insurance premiums. Mr Hyde exhibited to his witness statement an article from the Sydney Morning Herald, Latest low-speed crash tests favour large cars, dated 4 March 1999, in which it is stated that the 15 km/h pendulum tests were in their ninth year.

53. In cross-examination Mr Case was referred to the article in the Sydney Morning Herald and asked if he disagreed with what Mr Hyde had said about it. Mr Case said that he was concerned with the timing of the development of the test as he was not certain when it was developed. He said that he did not wish to commit to stating the time or date when the test was developed because he didn't have that knowledge. However, Mr Case said he did not dispute what was stated in the newspaper article.

54. The evidence before us at the hearing of this matter points to the fact that the low speed crash tests were developed by IAG well before the 1999 income year which is the first year in question in this matter. There was no evidence at all before us about any further test development conducted by IAG or by RACV Sales. The examples of crash testing results produced by RACV Sales between 1999 and 2006 make no mention whatsoever of the low speed crash test. This is despite the fact that Mr Case said in his witness statement that the low speed crash test system is tested and simulated on an ongoing basis on different vehicles to ensure that the hypothesis remains correct.

55. On the evidence before us on hearing this matter, we find that there was no activity which could properly be described as the development of low speed crash testing during the years in question. It was not a research and development activity conducted by or on behalf of RACV Sales between 1999 and 2006.

Activity 6: Continual development of side impact crash testing

56. This test is usually described as the mobile deformable barrier side impact crash test. In his witness statement dated 15 August 2011 Mr Paine said that ANCAP continually investigates and researches all aspects of crash testing. This general description does not identify the activities undertaken by RACV Sales.

57. In a paper prepared by Mr Paine which is dated 28 March 2004, he analysed the Potential Use of the Pole Test for Assessing Side Impact Performance. By way of background, Mr Paine said that ANCAP was looking at the introduction of the pole impact crash test, as that test was described in the EuroNCAP test protocol (Version 4.0), to replace the current EuroNCAP side impact test. He said nearly all vehicles scored well in the current side impact test and there appeared to be limited opportunity to spread the field with this type of test.

58. Mr Paine described the mobile barrier side impact test. He said that a 950 kg trolley with a crushable aluminium barrier is propelled into the side of the test vehicle at 50 km/h. A Euro SID II dummy, representing an average size adult male, is located in the driver's seat for this test and the risk of head, chest, abdomen and pelvic injury is determined from dummy sensors. Mr Paine explained that this test was based on a test conducted for the purposes of satisfying Australian Design Rule (ADR) 72, although the ADR was introduced after ANCAP commenced using this test. He said EuroNCAP started using this test in 1996. ANCAP commenced using the test in 1999 when it aligned its test protocols with those of EuroNCAP.

59. In cross-examination Mr Paine confirmed that EuroNCAP commenced using the side impact test in 1996 and that prior to harmonisation, ANCAP did not use the test. In fact, in order to bring about harmonisation with EuroNCAP, Mr Paine agreed that ANCAP ceased conducting its full frontal test and adopted the mobile barrier side impact test. Mr Paine also agreed that the test has remained the same, subject to variations in the protocol, since ANCAP commenced using it. Mr Paine said that the side impact test was the same test manufacturers were required to perform in order to obtain a compliance plate by complying with ADR 72.

60. The ANCAP Notes on the Assessment Protocol as at April 2011, which was an exhibit to Mr Paine's witness statement, states that ADR 72 was not applicable to high-seat vehicles. It stated that in 2004 ANCAP identified that there was little value to consumers in continuing to conduct the side impact test on high-seat vehicles in Australia and decided to award all high-seat vehicles a default score of 16 points for that test. Mr Paine confirmed that the ADR exempts vehicles with the seat height of 700 mm or greater from the regulation. Mr Paine also confirmed that in about 2004 when EuroNCAP introduced the Euro SID II dummy, in accordance with harmonisation, ANCAP also switched to that dummy. Mr Paine was then asked whether, except for the 16 point default score and the fact that the side impact test was not applied to vehicles with a seat-height exceeding 700 mm, any other change occurred with respect to the mobile impact test from 1999 until 2006. His response was_: no significant changes, no. Minor things like how the head restraints are adjusted and things like that. But only very minor stuff_.

61. In his witness statement dated 27 June 2011 Mr Case said that as part of considering crash test results from overseas and simulations in Australia, a review is made of real-world accident data from Australia to evaluate interpretations from crash testing and to ensure interpretations are made which simulate actual crash situations. He said it was discovered that by simply adopting the EuroNCAP mobile deformable barrier testing procedures, crash testing did not replicate real-world events and yielded higher safety observations than real-life empirical data.

62. Mr Case referred to an article entitled: Crash testing for safety-possible enhancements to ANCAP test and rating methods which was co-authored by Mr Paine. The introduction to the paper states that it was prepared for discussion purposes and should not be regarded as a policy statement. The paper refers to the fact that it became evident that the side impact test was not particularly demanding and most vehicles scored well in that test. Mr Case also described the problems experienced by ANCAP when performing the side impact test on vehicles with an elevated seat-height. He said the existing testing techniques were unsatisfactory for such vehicles as the impact boom from the side impact trolley had a prevalence to push underneath the sill panels, contacting the chassis rails and tipping the vehicle. This led to inconsistent data. A low barrier test was considered not appropriate for four-wheel-drive vehicles. According to Mr Case, this led to the development of side impact testing including the development of a new pole crash test.

63. In cross-examination Mr Case was referred to a statement he made in his witness statement after he explained the problems with the side impact test. Mr Case said: as a consequence, ANCAP has been developing side impact testing, including for four-wheel drives. He agreed that that was a reference to the pole test program.

64. Mr Case annexed to his supplementary witness statement dated 16 August 2011 a paper entitled: ANCAP Future Technical Directions, authored by Mr Christopher Coxon. The paper is not dated. It appears to have been written in about 2003. In his paper, Mr Coxon referred to a proposed speed increase for the side impact test. He also stated that EuroNCAP had made upgrades to use the Euro SID II dummy at the end of 2002 as well as changing the deformable face of the barrier to a progressive deformation construction. Mr Coxon then said that ANCAP would incorporate those changes in its side impact test during 2003.

65. In his witness statement Professor Zockel said that in the documents provided to him he had not found any records of investigations undertaken to determine options for modifying the side impact crash test. He also said he had not seen any records of experimental activities.

66. As Innovation Australia submitted, there was no evidence that RACV Sales undertook any developmental work regarding the side impact crash test. The adjective developmental means of or pertaining to development; evolutionary. The word development is defined in The Shorter Oxford English Dictionary as: a gradual unfolding; a full working out the details of anything. In our opinion, the most that can be said of the work done is that RACV Sales identified shortcomings when the side impact crash test was performed on vehicles with a seat height exceeding 700 mm. Those shortcomings were discovered in the course of its ordinary testing program. It was not the result of any innovative research conducted by RACV Sales for the purpose of developing the side impact crash test.

67. While we have no doubt that the shortcomings identified in using the side impact crash test led RACV Sales to substituting the pole test for the side impact crash test, we are unable to discern, from the evidence before us, any activity of a research and developmental nature, or of any kind at all which can be described as a development of the side impact crash test. We therefore find that the activity described by RACV Sales as continual development of side impact crash testing does not satisfy the definition of research and development activities for the purposes of

68. The evidence regarding development of a pole crash test was contradictory. In his witness statement dated 27 June 2011 Mr Case said that RACV Sales identified the need for a pole crash test because the side impact test did not replicate real-world injury outcomes when conducted on four-wheel drive vehicles. He said the hypothesis adopted by ANCAP was that head-protecting side impact airbags would generate positive safety outcomes. This statement does not sit well with what is stated in a paper co-authored by Mr Paine entitled: Consumer crash tests: the elusive best practice. This paper was presented at a symposium in December 1999. The paper states that there is a provision in the EuroNCAP protocol for a side impact pole test to be conducted at the manufacturer's expense. It was said to apply where a maximum head score is achieved in the side impact barrier test and a head protecting side airbag was provided. The pole test was said to earn the vehicle an additional two points and, if the total score exceeded 32 points, a fifth star.

69. In a paper prepared by Mr Paine in March 2004 entitled: Potential Use of Pole Test for Assessing Side Impact Performance, he stated that ANCAP was looking at the introduction of the pole impact crash test as described in the EuroNCAP test protocol to replace the current EuroNCAP side impact test. He referred to the fact that some new cars at that time had a new head protection device, like a curtain airbag. Therefore, pole impact and other tests should be introduced to evaluate these kinds of devices and head injury risk. Mr Paine also explained that the pole test vehicle travelled sideways at 29 km/h and impacted a fixed pole of 300 mm diameter which was lined up with the driver's head. He said that the same injury measurements as those for the side impact test are required to be recorded although only the head injury measurements are reported and taken into account in the assessment. Mr Paine then said that ANCAP wished to determine the likely effect of scoring the pole test in the same way as the side impact test.

70. Mr Case said in his witness statement that ANCAP needed to develop a process, hardware and rig to meet the features of the Australian fleet, which involved smaller lighter vehicles on average, with different safety features to those found in the United States of America. He then said that the pole crash test developed by ANCAP broadly followed the EuroNCAP protocols and involved propelling the vehicle sideways at 29 km/h into a narrow rigid pole. He said that the hardware and rig to incorporate standard pole testing of this type to four-wheel drive vehicles did not exist in Australia and had to be developed by ANCAP. Mr Case referred to answers given to questions posed by Innovation Australia regarding the development of a pole crash test where it was said that the hardware and test rig currently did not exist and that RACV Sales would be developing this testing method and equipment. He also said that a three stage process of analysing the effectiveness of the rig and hardware and to trial the testing methodology was developed.

71. In a discussion paper prepared by Mr Paine which is dated 19 January 2001, he explained the EuroNCAP optional pole test. He said that at that time about eight pole impact tests had been conducted in Europe. No pole impact tests had been conducted in Australia. He also explained that although a fully instrumented Euro-SID I dummy was used in the pole test, only the head acceleration was currently used in the assessment. He said the relevance of the other injury measurements needed to be assessed. Mr Paine also said: no pole tests have been conducted in Australia, although Crashlab has the capability to conduct such tests. Crashlab is one of the crash testing laboratories utilised by ANCAP.

72. In another paper co-authored by Mr Paine which is undated but appears to have been written in about 2001, the authors state that no vehicle tested by Australian NCAP has been eligible for the pole test. The authors state that pole tests were conducted according to the EuroNCAP protocol in Europe. They also state that head, chest and abdomen and pelvis injury measurements are recorded on the driver dummy.

73. In cross-examination Mr McCarthy put to Mr Case that when he talked about the pole crash test developed by ANCAP, he didn't mean the test itself, because that was already in place, but rather about a program to find out how effective that would be on four-wheel drive vehicles. Mr Case responded_: not the test itself, or its protocol, but about its application_. Mr Case confirmed that what RACV Sales was looking into was how useful that test might be on four-wheel drive vehicles. He also confirmed that while RACV Sales considered adopting a scoring system which differed from EuroNCAP, that did not occur because the protocol remained the same. When it was put to Mr Case that exploring the possibility of analysing and validating appropriate test protocols was something which was common with all tests in order to determine whether they were providing useful information, he agreed.

74. In cross-examination, Mr Paine confirmed that the pole test was first done by EuroNCAP 1999. ANCAP did its first pole test in 2003. He also confirmed that EuroNCAP was concerned that the side impact test did not really challenge the protection system for head injury and that concern was universal. Mr Paine also agreed that ANCAP, as result of the memorandum of understanding which led to harmonisation, needed to adopt the pole test where Australian vehicles were fitted with side protecting airbags. He agreed that the test itself, in terms of what the test actually is, remained constant at all times.

75. Despite what Mr Paine said in the paper he wrote in 2001 where he referred to Crashlab having the capability to conduct the pole test, in cross-examination he agreed that ANCAP did not have the facilities to do a pole test in the early 2000s. He did however agree that in due course facilities did become available. He said that when the first test was conducted for ANCAP in Japan, an engineer from a local test organisation went with him to look at the way in which that test was conducted and that was part of the process of developing a trolley on which the motor vehicle was placed to enable it to be propelled into the fixed pole. Mr Paine also confirmed that ANCAP was anxious to make sure that its test was conducted with sufficient precision and quality control so that it complied with the protocol.

76. Given the prevalence in Australia of four-wheel drive vehicles with a seat height exceeding 700 mm, Mr Paine said it occurred to him that using the pole test might be a good way to record information not simply about head injuries, but injuries which occurred to other parts of the body in such accidents. This is what prompted him to produce a discussion paper in 2001. Although in his discussion paper Mr Paine suggested that the pole test could be substituted for the side impact test, he agreed that this suggestion did not proceed although ANCAP decided to make it a requirement that vehicles needed to score at least one point on the pole test for a five-star rating. That decision was made in about 2004.

77. In re-examination Ms Baker asked Mr Paine to describe the non-routine part of ANCAP's work. Mr Paine said the non-routine work included:

(a) identifying vehicles that should be crash tested or results that should be republished from EuroNCAP, in other words, keeping an eye on the Australian car market;

(b) coming up with a short list of vehicles that should be considered for crash testing;

(c) arranging crash tests, acquiring vehicles and organising the test labs;

(d) attending the Crashlab immediately before a test to make sure that everything was in accordance with the protocol and that there were no problems;

(e) obtaining injury data and test lab reports of intrusion data and other observations looking for modifiers (reasons to deduct points from scores from injury scores to reflect injury hazards not picked up in the dummy measurements);

(f) quality testing to make sure labs have followed protocols; and

78. Ms Baker asked Mr Paine whether any of the information or knowledge that came from the testing was new. His answer was that all of the steps were fairly routine and straightforward as he had developed forms to try and make them as objective and quantitative as possible.

79. The description used by RACV Sales regarding this activity is, in our opinion, misleading. There was no evidence before us whatsoever of any developmental activity undertaken regarding the pole crash test itself. For example, there was no evidence of the testing of various sizes of poles or speeds at which the trolley carrying the motor vehicle was directed at the pole. Nor was there any evidence of testing various positions at which the motor vehicle would strike the pole, other than ensuring that it was in line with the head of the driver of the vehicle. In fact, in cross-examination, Mr Case admitted as much stating that the development was not of the crash test itself or the protocols but rather the application of the test. By that statement, we understood Mr Case to be stating that RACV Sales simply applied the test to a variety of vehicles. While we have no doubt that the initial testing conducted by RACV Sales on four-wheel drive vehicles of the kind commonly used in Australia would have commenced with some degree of uncertainty about the validity or usefulness of test results, the process of conducting those tests was not novel nor was it new. The EuroNCAP protocols for conducting those tests were established and followed by ANCAP.

80. Although Mr Case referred to the development of the hardware and test rig, in cross-examination he made it clear that RACV Sales was not involved in the development of the hardware or the rig. In fact, he agreed that what RACV Sales did was to simply develop programs for the testing of vehicles using established hardware and protocols. In his 2001 discussion paper, Mr Paine stated that the capability to conduct a pole crash test already existed in Australia. There was no evidence before us at all that RACV Sales modified or adapted that test either prior to or after commencing its testing of four-wheel drive vehicles. All that occurred was an existing test process was applied to a different style of motor vehicle and the results obtained from that testing were compared with real-world data in order to evaluate the usefulness of the results. There is nothing in this activity which can be described as innovative or involving high levels of technical risk. Activity 7 does not have any feature which is new. There is no chance that the process involved in this activity will fail to work or achieve an outcome which was unintended. The measurements taken from the testing are what they are. We therefore find that this activity does not satisfy the definition of research and development activities set out in ITAA 1936.

Activity 8: Development of pedestrian crash test

81. In his witness statement Mr Case said that ANCAP was the first to apply a pedestrian impact test in Australia in 2000 based on testing protocols developed by EuroNCAP. The crash test involved measuring outcomes on pedestrians of crashes up to 40 km/per hour. Mr Case said that in developing the pedestrian crash test, ANCAP was required to consider real-life accident events using data from ANCIS to develop testing techniques which would identify potential design areas to enhance safety; model and test the protocols; enter into a pilot program; analyse results and modify the proposed testing method; implement the crash test; and continually correlate it to real-world events to ensure accuracy.

82. Mr Case said that ANCAP also identified the need to improve the pedestrian crash test to ensure it took into account common high riding vehicle specifications from the Australian fleet and modifications which are prevalent in Australia and not overseas, such as bull bars. He said there had been ongoing analysis to ensure ANCAP continued to co-ordinate this test with EuroNCAP protocols which is required as part of the memorandum of understanding.

83. In his witness statement Mr Paine said that EuroNCAP commenced pedestrian tests in 1997. In a paper co-authored by Mr Paine which was presented at the Impact Biomechanics and Neck Injury 2000 Institution of the Engineers Australia, it is stated that during 1999 ANCAP aligned its crash test and assessment procedures with those of EuroNCAP, including the EEVC. The authors also stated that the EEVC procedures were now used by ANCAP. The procedures required determination of locations on the vehicle that are most likely to be injurious and those that are least likely to be injurious. The vehicle manufacturer may nominate the locations on the vehicle that are least likely to be injurious provided that they are not too close to locations judged to be injurious. The EEVC procedures require three sub-system tests that simulate impacts by the head, upper leg and knee/lower leg.

84. In the application forms lodged by RACV Sales for the 2005 and 2006 income years, the description of the test remained identical to previous applications but, in addition, the applicant said:

"It is proposed that for the 2005/06, RACV Sales will undertake further development with respect to the pedestrian impact test. These tests have to date been based on overseas programs. Further test development is required for local conditions, for example incorporation of pedestrian tests for higher vehicles and vehicles fitted with bull bars."

85. When it was put to Mr Case in cross-examination that where he referred to developing the pedestrian crash test in his witness statement, what he meant was taking the pedestrian crash test from the EuroNCAP program, adopting it under the protocol and causing it to become part of the ANCAP program, Mr Case responded_: broadly, yes_. Mr Case added that it did involve working with the contractor to develop and build the rig and test it to make sure that it was going to deliver the same, or the same test according to the protocol. In re-examination, Mr Case was asked to describe what he meant by ensuring that the pedestrian test also applied in Australia. He said that as far as ANCAP was concerned, it was about establishing the suitability of the pedestrian test in Australian conditions. That meant building a test rig to EuroNCAP specifications and trialling it on locally produced vehicles to see whether the test results would provide useful advice to consumers about the effects of their vehicles on impacts with pedestrians. However, despite Mr Case's oral evidence, there was no objective evidence before us of RACV Sales working with a contractor to develop, build and test the rig. In questions from the Tribunal Mr Case agreed that the pedestrian test was exactly the same test used by EuroNCAP.

86. In cross-examination, Mr Paine confirmed that as part of the harmonisation process introduced in 1999, ANCAP adopted exactly the same pedestrian protocols as those used by EuroNCAP. ANCAP uses exactly the same rating systems as EuroNCAP and the test is conducted in exactly the same way. Mr Paine said that the test was the same up until 2006 and there was a change shortly after that. He also agreed that ANCAP had a slightly different system for giving star ratings as it provided a separate rating for pedestrian protection. However, that was in accordance with the memorandum of understanding.

87. In his witness statement Mr Hyde stated that Europe is the world leader in the area of pedestrian impact tests and applying the results of that testing to produce good design. He said this could be seen in many European designed vehicles which feature high engine hoods which are designed to minimise impact forces and therefore injury to the pedestrian. Mr Hyde referred to the description of EuroNCAP's pedestrian testing which is identical to that adopted by ANCAP. He said he was unable to identify any difference between the pedestrian testing described as being carried out by ANCAP and the tests conducted overseas by EuroNCAP.

88. In his witness statement Professor Zockel said, after reviewing the material which was made available to him, that he could not identify any details of which experiments were performed or how any such experiments led to the development of the pedestrian test. Although Professor Zockel said he was aware of research conducted by the Centre for Automotive Safety Research, which is a research centre at the University of Adelaide, he said he was not aware of the details of that research. He had not seen any document evidencing the work which was undertaken and therefore said he was unable to express an opinion on whether it involved the development of a new test, or, if so, whether it involved innovation or technical risk.

89. The evidence before us on this particular matter does not disclose innovation or technical risk. The pedestrian test performed by ANCAP is identical to that developed by EuroNCAP by at least 1999. Although in the cross-examination of Professor Zockel, Ms Baker attempted to obtain from him that if the results from real-world crashes outside of any crash testing program produced results which were different from those which were predicted through crash testing, that involved technical risk, we do not agree. While we have no doubt that all of ANCAP's crash testing results are matched against information obtained from real-world crashes, the purpose of this matching exercise is simply to confirm the validity of the testing conducted. As Mr Paine said in his witness statement, RACV Sales had compared the results of ANCAP tests with the outcome in real-world crashes and those comparisons disclosed a reasonable correlation between ANCAP's crash testing, the assessment results and real-world crash data.

90. Mr Paine said in cross-examination that some of the papers which he co-authored were written for the purpose of reviewing what was happening around the world regarding crash testing and to identify problems which had been experienced using EuroNCAP protocols. There is nothing in the evidence which suggests that the invalidity of any test results led to ANCAP conducting experiments with its testing procedure in an attempt to validate its crash test results. The evidence regarding pedestrian crash testing does not disclose systematic, investigative and experimental activities which involve innovation or high levels of technical risk. Accordingly, we find that this activity does not satisfy the definition of research and development activity as described in s 73B(1) of ITAA36.

Activity 9: Continual development of key information presentations

91. In his witness statement Mr Case said that while overseas protocols, particularly those from EuroNCAP, are generally used, divergences could occur where data from actual crash situations led to differences in crash testing, and such crash testing led to different results. He said that concentration on ANCAP injury content was considered too obscure for the intended audience and simplified formats were required which retained the integrity of the data and were internationally recognised. He said techniques were redesigned depending on comparisons with real-world data in order to achieve optimum consumer comprehension. The addition of technical reports and colour-coded human figures to indicate the extent of head and chest injuries are examples of the innovative formats and techniques resulting from this activity.

92. Mr Case's witness statement referred to a paper, Harmonisation of Australian NCAP with EuroNCAP - Lessons Learned, which he co-authored with Mr Jack Haley of NRMA Insurance and Mr Paine. Although the paper is undated and Mr Case has not indicated the date of writing that paper, it appears to have been written shortly after 2000. The authors state that the paper addresses several concerns about the EuroNCAP protocols and it provides constructive suggestions for changes to that protocol. The authors state that the ANCAP Technical Committee members were reviewing the operation of the EuroNCAP protocols since their adoption by ANCAP and have made representations to the EuroNCAP Technical Chair in relation to desired changes. This indicates, in our opinion, that ANCAP followed EuroNCAP protocols although when confronted with results which did not necessarily reflect the risk of serious injury, it raised its concerns with EuroNCAP. It did not act unilaterally to change either the tests or the way in which those results were presented. Rather, it made recommendations about the changes it thought would more accurately reflect the risk of injury. For example, the authors state that the EuroNCAP lower limit of 22 mm for chest compression may be unrealistically low. It probably did not reflect the risk of serious chest injury. The authors recommended that the lower limit of chest compression of 30 mm and the upper limit of 50 mm be retained.

93. As far as the overall rating of a motor vehicle on a crash test is concerned, the authors of the paper stated that, due to the equal weight given to offset and side impact scores in the overall score and a lack of modifiers for the side impact test, vehicles could earn a three star rating with a very poor offset score. They also stated that a similar situation occurred when the vehicle was eligible for the optional pole test and it earned an extra two points. In this case, the side impact and pole tests can add up to 18 points and the vehicle need only earn 7.5 points in the offset test to reach four stars. The authors of the paper recommended that the star rating system be amended to include cut-off values for the offset score.

94. In his supplementary witness statement Mr Case stated that there was no requirement to harmonise the results presentation format as part of the EuroNCAP and ANCAP harmonisation process. It was always intended that each country would choose the presentation format that best suited its home market. However, in a paper referred to by Mr Case, which he co-authored, entitled Evolution of Australian NCAP Results Presentation, the authors state that EuroNCAP began publishing its offset and side impact test data in February 1997 using a star rating system. They said that ANCAP was working with EuroNCAP and other international NCAP groups towards a common rating system. At the time of writing this paper, ANCAP had not adopted a star rating system but rather used the good, acceptable, marginal and poor ratings. The authors of the paper stated that ANCAP had commissioned a focus group to research and evaluate how well the current brochure designs were communicating the ANCAP results. Its purpose was to canvass how the design and distribution could be improved and to suggest other avenues for promoting the program and the information.

95. In cross-examination Mr Case agreed that when it came to giving star ratings, the first issue was to ensure that all agencies which gave star ratings to motor vehicles on crash testing used the same star ratings. Where ANCAP did something which was different to how EuroNCAP did it, that was reflected in the protocol.

96. We accept that since 1999 ANCAP has periodically reviewed and altered the methods of presentation of information to the car buying public. Furthermore, this stated activity clearly overlaps with Activity 3 which deals with the continual development of data interpretation and presentation techniques. The process adopted by ANCAP is properly described as one of simplifying the presentation of complex statistical data. However, it cannot be said that, by distilling the data such that it can be presented in the form of a star rating or the presentation by use of coloured figures representing drivers and passengers indicating the risk of injury to various body regions, the activity can properly be described as involving innovation or high levels of technical risk. It is properly described as a deductive process whereby the measurements obtained in the course of crash testing are reproduced in a simple format enabling the consumer to obtain a general understanding of the risk of serious injury in a particular vehicle should that vehicle be involved in a crash. We find it does not satisfy the definition of research and development activities as that expression is defined in ITAA36.

Activity 12: Continual design of testing procedure and data manipulation

97. Once again, this particular activity appears to overlap to a considerable degree with a number of the other activities to which we have referred above. On its application forms, RACV Sales described the program as incorporation of new testing techniques to assess car safety and new tests performed which assist in identifying car safety in side impact and pedestrian impact.

98. There was no evidence before us at all that ANCAP was involved in the design of any testing procedure. In fact the evidence discloses that ANCAP adopted the procedures designed by EuroNCAP which is set out in its testing protocols. In a paper authored by Mr Coxon, ANCAP Future Technical Directions, Mr Coxon describes the international harmonisation of tests. He stated:

"Where possible, ANCAP will continue to use the best international NCAP type tests for the basis of the Australian and New Zealand NCAP. This allows our small resources to be coupled to larger organisations that develop testing and assessment protocols. International manufacturers are also supportive of this approach as it is not an ANCAP intention to carry out unique tests unless there is a compelling safety and technical reason. This has been the practice that has given Australian new car buyers information to over 75% of the volume new vehicles sold in the Australian market on individual vehicle occupant protection information."

99. In his witness statement Mr Case referred to ANCAP developing the pedestrian crash test. We have already referred to Mr Case's statement that ANCAP was required to develop testing techniques. However, as Mr Paine said in his witness statement, the test techniques used by ANCAP are those devised by the EEVC. ANCAP has aligned its pedestrian crash test and assessment procedures with those of EuroNCAP. In a paper entitled Assessment of Pedestrian Protection Afforded by Vehicles in Australia, which Mr Paine co-authored with Mr Coxon and presented to a seminar in March 2000, the authors stated that consumer tests conducted by EuroNCAP revealed there was room for a great deal of improvement with current vehicle models when it came to improving survivability for pedestrians struck by motor vehicles. The authors stated that the situation was likely to be similar in Australia, where ANCAP has recently commenced testing to the EuroNCAP pedestrian protection protocol.

100. There was no evidence before us of new testing techniques having been developed by ANCAP or of any new tests performed to assist in identifying car safety in side impact and pedestrian safety in pedestrian crash tests. This is despite the fact that, as Mr McCarthy said in his opening remarks, each of the application forms lodged by RACV Sales contains a statement whereby the maker declared that he or she was authorised by the company to make the declaration and that the company had, while carrying on the activities, maintained contemporaneous records which substantiate the company's carrying on of the activities. While work has clearly been done on improving the data presented to consumers, the evidence does not disclose that this work is either innovative or that it involves high levels of technical risk. We find that it does not satisfy the definition of research and development activities as that expression is used in ITAA36.

101. This activity is said to include a number of sub-programs. They are described as the small car program, the large car program, the 4WD program (formerly the utility program), roll over car tests, and electronic stability control (ESC) testing. However, like many of the other programs, we did not have any evidence in the form of contemporaneous records of experiments or other activity.

102. In its application forms for the 2001 and 2003 income years, RACV Sales described the small car program as the first stage of ANCAP testing designed and developed to allow harmonisation with EuroNCAP testing procedures. RACV Sales submitted that the technical committee learnt valuable lessons from the first and subsequent launches of this program, for example, that crash tests cannot be directly transferred from European conditions to Australian conditions. Although it was stated that modifications needed to be made to the tests and the analysis to allow for factors such as differences in passenger restraint systems between Australian and European models, we had no direct evidence of modifications made to the tests conducted by ANCAP. The only significant difference between ANCAP testing and tests conducted by EuroNCAP which resulted from the prevalence of high seated vehicles in Australia was the mandatory pole test required by ANCAP for a vehicle to obtain a five-star rating. That did not require ANCAP to develop the pole test because the test had already been devised and protocols developed by EuroNCAP in about 1999.

103. In a paper co-authored by Mr Paine and Mr Haley, Crash Testing for Safety-Possible Enhancements to ANCAP Test Rating Methods, which appears to have been published in about November 2008, the authors set out the differences between EuroNCAP and ANCAP protocols. In essence, the differences are about the points attributed to various tests. They are:

(a) for a five-star rating, at least one point must be scored in the pole test;

(b) for a five-star rating, the tested model must have ESC available either as standard or an option;

(c) vehicles with high seating positions are not subjected to a side impact test and instead a default score of 16 points is awarded;

(d) ANCAP applies a points balance system, for example, if a vehicle scores less than 12.5 points in the offset test it is not eligible for five stars;

(e) the assessment of knee modifiers is slightly different to that used by EuroNCAP;

(f) where EuroNCAP tests a model with a driver knee airbag and this is not available in Australia, two points are deducted from the upper leg score unless there is evidence to show that this is inappropriate;

(g) steering wheel and pedal intrusion are measured relative to the final position of the driver's seat; and

(h) ANCAP includes child dummies in child restraints for the offset and side impact tests but does not currently assess child occupant protection.

104. Although RACV Sales in its application forms stated that modifications needed to be made to the tests and the analysis, there was no evidence at all of any modification to the tests although, as described above, quite plainly there were modifications to the analysis. The modifications to the analysis appear to have been made so that the results more closely conform to the results obtained by MURAC of real-world crashes. Those modifications cannot be described as innovative or involving high levels of technical risk.

105. According to RACV Sales, the large car program resulted from significant occupant protection upgrades across all cars within the large car market. Mr Case referred to the development of the anti-lock braking system (ABS). Those upgrades needed to be assessed by ANCAP and analysed. RACV Sales claimed that often the testing and analysis equipment needed to be reconfigured to adapt to new testing parameters arising from new car components. Once again, the problem for us is that there was no evidence before us of any reconfiguration of testing equipment or of any new testing parameters arising from new car components in the 2001 income year.

106. In its application form RACV Sales claimed that a new testing procedure had been designed for utilities. We had no evidence before us of this new procedure. RACV Sales claimed that this testing would also be used to investigate the effect of sill height on the response of cars in a side impact test. We assume that this is a reference to the pole test. Again, there is nothing new or innovative about the pole test.

107. In their paper published about November 2008, Crash Testing for Safety - Possible Enhancements to ANCAP Test and Rating Methods, Mr Paine and Mr Haley said that a major concern in Australia and New Zealand was the commercial vehicle market, which included utilities, pick-ups and vans. The authors stated that several models did not have a driver airbag and that about two thirds of commercial vehicle sales were vehicles rated at three stars or less. Despite expressing those concerns, the authors make no mention whatsoever in their paper about any new testing for this type of vehicle. Nor was there any evidence from any witness about the development of new testing for this type of vehicle.

108. In its application form for the 2004 income year, RACV Sales claimed that it would undertake the development of a new locally developed rollover test. There was no evidence before us of any development of this test.

109. In its application forms for the 2005 and 2006 income years, RACV Sales said the rollover car test and electronic stability control testing was proposed as a new activity. It stated that currently the development was in its infancy and that further bases for the tests would follow in the 2005 year and beyond. Curiously, in the paper co-authored by Mr Paine in 2008, while the authors deal with future directions regarding crash testing, no mention is made of the development of a rollover crash test. The authors only mention that ANCAP will most likely need to review several rating-related issues including ways to assess active and passive safety in a rollover crash. This suggests to us that rollover crash testing had not commenced by 2008.

110. The authors also state that ANCAP started to require ESC for a five-star rating from January 2008. No mention is made of any testing undertaken to establish the enhancements to the safety of motor vehicles which were equipped with ESC. In his witness statement Mr Paine said that around 2005, he started work researching ESC and began investigating how it could be brought into ANCAP's rating system to encourage vehicle manufacturers to introduce it in Australia. Mr Paine wrote a paper in June 2005 dealing with ESC (Electronic Stability Control: Review of Research and Regulations). This paper is, in effect, a literature review of the subject. Mr Paine said in his paper that during February and March 2005 key overseas researchers were contacted for advice about the status of research projects on ESC and that those researchers were looking at ways to assess the performance of ESC. Mr Paine referred to the US National Highway Traffic Safety Administration (NHTSA) which was, at that time, investigating possible dynamic tests for determining the performance of ESC and that the development work was still under way.

111. In his paper Mr Paine also examined what EuroNCAP had been doing in relation to the fitment of ESC systems by manufacturers to motor vehicles. He referred to a paper written by a Mr Williams from EuroNCAP who said that none of the research that EuroNCAP had seen was able to identify why ESC systems were so effective save for the fact that they simply found good correlation between the fitment of an ESC system and a reduction in the number of accidents. He said it posed a problem when trying to develop a test of ESC systems. He also said that EuroNCAP could start simply by recommending that consumers buy cars with ESC. A step-up from this would be to award points to motor vehicles fitted with ESC. EuroNCAP might then want to develop a test which simply checks that the ESC system works.

112. In concluding, Mr Paine recommended that Australian authorities monitor overseas developments concerning the assessment of ESC. He said that at this stage, none of the surveyed overseas projects had produced a performance-based test protocol and/or rating system that was fully suitable for use in a consumer programme. Mr Paine also recommended that a short research project be funded to evaluate the potential for an Australian-developed handling test to be used for assessing ESC. There was no evidence before us that the short research project recommended by Mr Paine had commenced by the end of the 2006 income year.

113. Given the evidence which was available to us at the hearing of this matter regarding roll-over testing and the use of ESC, we find that the claimed activities regarding the development of roll-over testing and the testing of ESC had not proceeded at all in the income years in question and therefore cannot satisfy the definition of research and development activities as that expression is defined in ITAA36.

Activities 1, 2, 4, 10, 11 and 14: The claimed directly related activities

114. The remainder of the activities to which we have referred in the table set out in paragraph [7] above are said to be directly related to the systematic, investigative and experimental activities which we have discussed in some detail above. They are activities 1, 2, 4, 10, 11 and 14. As we have found that all of the activities described by RACV Sales as systematic, investigative and experimental do not fit that description, the activities described as directly related cannot satisfy the definition of research and development activities as that expression is defined in the 1936 Act.

115. Despite careful review of the activities undertaken in relation to the projects referred to by RACV Sales, we were unable to identify any activities which satisfy the definition of research and development activities in s 73B(1) of ITAA36. RACV Sales' involvement through ANCAP in the activities which we have analysed above and which were described in its application forms for registration of research and development activities cannot properly be described as systematic, investigative and experimental activities that involve innovation or high levels of technical risk. The various references to development of crash testing activities is, with respect, misleading. All of the tests conducted by RACV Sales through ANCAP had been previously developed by other organisations. There was no evidence that ANCAP conducted any developmental work regarding the tests themselves.

116. In fact, the very use of the word "test" by RACV Sales appears to be inappropriate. The Shorter Oxford English Dictionary defines the word test, in relation to mechanics as: "the action by which the physical properties of substances, materials, machines, etc. are tested, in order to determine their ability to satisfy particular requirements." ANCAP conducts tests on various types of motor vehicles to assess the risk of injury to its occupants, or to pedestrians struck by them. It does not test those vehicles with a view to determine their ability to satisfy any particular requirement. MUARC, which conducts an analysis of real-world crashes, describes the process as the measurement of crashworthiness and aggressivity of the various models of motor vehicles involved in motor vehicle crashes. A crashworthiness rating is given by the product of injury, of risk and injury severity. The work done by ANCAP, although involving a controlled crash, is not dissimilar to the work conducted by MUARC and is more accurately described as that of data collection and assessment. Mr Paine described the process of data collection as routine work. He also said that he had developed forms setting out routine steps to be taken in the acquisition of the information from crash testing.

117. RACV Sales also contended that innovation may reside in new knowledge obtained and introduced through the activities or new products in terms of meaningful crash test ratings. Ms Baker submitted that the activities conducted by ANCAP were directed at applying test procedures in Australia for the first time. Systematic steps were taken to develop crash testing, assessment and rating protocols that were specifically directed to and appropriate for the unique vehicle mix and conditions in Australia.

118. Although we had evidence of the fact that the value of tests conducted by EuroNCAP could not always be translated into meaningful data as far as Australian vehicles are concerned, as Mr McCarthy submitted, the results of the tests are what they are. Furthermore, the results of the tests are produced by ANCAP's application of EuroNCAP protocols to the testing procedure. The scores obtained from various measuring devices in the motor vehicle crash tested and the dummy installed in that vehicle are put through a pre-existing computer program to determine the risk of serious injury to the occupants of the vehicle or, in pedestrian crash testing, the pedestrian. The process of the so-called crash testing and the measurements taken in the course of that process cannot properly be described as innovative.

119. Ms Baker submitted that the activities carried on by ANCAP were attended by a number of risks of a technical nature. These included:

(a) uncertainty as to whether the crash tests themselves would produce unintended outcomes, such as encouraging increased aggressivity of vehicles;

(b) uncertainty that what was tried would succeed in developing meaningful ratings for vehicles that correlate with real-world crash data;

(c) whether the activities enabled the crash test ratings to be presented in such a way that would result in improved vehicle design for occupant and pedestrian safety; and

(d) the way in which each vehicle tested would perform in the crash test was unknown and could not be determined in advance.

120. As we have indicated above, section 70B(2B)(b) must be considered when examining the definition of research and development activities. Activities are not taken to involve high levels of technical risk unless the probability of obtaining the technical or scientific outcome of the activities cannot be known or determined in advance; and the uncertainty of obtaining the outcome can be removed only through a program of systematic, investigative and experimental activities in which the scientific method has been applied. We have referred to the definition of the word "experiment" which means a trial carried out in order to test the theory, a machine's performance or to discover something unknown.

121. While it may be correct to say that the crash testing conducted by ANCAP might produce uncertain results as the outcome will not be known in advance, the crash testing is not an experiment as that word is used in s 73B of ITAA36. Its purpose is not to test a theory or the motor vehicle's performance or to discover something which is unknown. Its purpose is simply to make an assessment in each individual vehicle's case of the risk and severity of injury to the occupant or a pedestrian struck by the vehicle. Although it may also be said that at times the crash testing may produce an unexpected outcome, that is not its purpose.

122. Although Ms Baker submitted that the purpose of crash testing was to obtain data which was consistent with the findings of crash analysis in real-world crashes, that is not evident from the materials before us on the hearing of this matter. The purpose of comparing the data obtained from crash testing with real-world crash data is simply to verify the validity of the procedures adopted by ANCAP and, possibly, the protocols established by EuroNCAP.

123. As Mr McCarthy submitted, the presentation of the data obtained from crash testing is a policy decision. It is about providing consumers with simplified information to enable them to make a decision about the safety of the motor vehicle which they may be intending to purchase. By influencing consumer choice, ANCAP aims to put pressure on manufacturers to improve motor vehicle design. It is also about informing manufacturers, if they have not already conducted the testing themselves, of any unforeseen problems with their motor vehicles in the event of a crash. While it is hoped that manufacturers of motor vehicles might be moved by the outcome of crash testing, there is no guarantee that would be the case. That is not its purpose.

124. It follows from the reasons we have given, that we have decided that Innovation Australia is correct in deciding that RACV Sales' activities are not research and development activities as those activities are defined in s 73B(1) of ITAA36. Therefore, we have decided to affirm its decision to issue a certificate under s 39L of the IRD Act to that effect.

125. With effect from 29 January 2001, s 73B(1AAA) was inserted in ITAA36.[27] It set out Parliament's object in providing for a deduction in respect of research and development activities. It provides:

"The object of this section is to provide a tax incentive, in the form of a deduction, to[[28]] make eligible companies more internationally competitive by:

(a) encouraging the development by eligible companies of innovative products, processes and services; and

(b) increasing investment by eligible companies in defined research and development activities; and

(c) promoting the technological advancement of eligible companies through a focus on innovation or[[29]] high technical risk in defined research and development activities; and

(d) encouraging the use by eligible companies of strategic research and development planning; and

(e) creating an environment that is conducive to increased commercialisation of new processes and product technologies developed by eligible companies.

The benefits of the tax incentive are targeted by being limited to particular expenditure on certain defined activities."

An "eligible company" is "... a body corporate incorporated under a law of the Commonwealth or of a State or Territory."[30]

126. Various deductions are allowed under s 73B. Relevant in this case is that allowed under s 73B(14) which provides:

(a) an eligible company incurs research and development expenditure (other than contracted expenditure) during a year of income; and

(b) the aggregate research and development amount in relation to the company in relation to the year of income is greater than $20,000,

the amount of that expenditure multiplied by 1.25 is allowable as a deduction from the assessable income of the company of the year of income.

127. The expression "research and development expenditure" is defined in s 73B(1) in terms that include "... expenditure ... incurred by the company during the year of income, being ... other expenditure incurred on or after 1 July 1985 directly in respect of research and development activities carried on by or on behalf of the company on or after 1 July 1985 ...".[31]

Research and development activities under ITAA36

128. The expression "research and development activities" means:

"(a) systematic, investigative and experimental activities that involve innovation or high levels of technical risk and are carried on for the purpose of:

(i) acquiring new knowledge (whether or not that knowledge will have a specific practical application); or

(ii) creating new or improved materials, products, devices, processes or services; or

(b) other activities that are carried on for a purpose directly related to the carrying on of activities of the kind referred to in paragraph (a)."[32]

B. Activities removed from the scope of the definition

129. Unless activities are carried on in accordance with a plan that complies with guidelines formulated by the Board under s 39KA of the IRD Act and in force at the relevant time, activities are not covered by the definition.[33] It matters not that the activities meet all the criteria in the definition. That is the effect of s 73B(2BA) of ITAA36.

C. Criteria in definition qualified by other provisions

C.1 "Systematic, investigative and experimental activities"

130. Even if activities might be regarded as "systematic, investigative and experimental activities" for the purposes of the definition of "research and development activities" in s 73B(1), they will not be regarded as such if they are activities identified in s 73B(2A) or (2C). Section 73B(2A) is concerned with the development of computer software and is not relevant.[34] Although only ss 73B(2C)(a), (d), (i) and (n) may have any relevance in this case, we set the provision in its entirety as it becomes relevant in our interpretation of the expression "systematic, investigative and experimental activities".[35] Section 73B(2C) provides that, for the purposes of s 73B:[36]

"... the following activities are taken not to be systematic, investigative and experimental activities:

(a) market research, market testing or market development, or sales promotion (including consumer surveys);

(c) prospecting, exploring or drilling for minerals, petroleum or natural gas for the purpose of discovering deposits, determining more precisely the location of deposits or determining the size or quality of deposits;

(d) the making of cosmetic modifications or stylistic changes to products, processes or production methods;

(f) research in social sciences, arts or humanities;

(h) pre-production activities such as demonstration of commercial viability, tooling-up and trial runs;

(i) routine collection of information, except as part of the research and development process;

(k) commercial, legal and administrative aspects of patenting, licensing or other activities;

(n) any activity related to the reproduction of a commercial product or process by a physical examination of an existing system or from plans, blueprints, detailed specifications or publicly available information."

C.2 "innovation or high levels of technical risk"

131. The definition of "research and development activities" must be read in light of s 73B(2B). It provides:

"For the purposes of the definition of research and development activities in subsection (1):

(a) activities are not taken to involve innovation unless they involve an appreciable element of novelty; and

(b) activities are not taken to involve high levels of technical risk unless:

(i) the probability of obtaining the technical or scientific outcome of the activities cannot be known or determined in advance on the basis of current knowledge or experience; and

(ii) the uncertainty of obtaining the outcome can be removed only through a program of systematic, investigative and experimental activities in which a scientific method has been applied, in a systematic progression of work (based on principles of physical, biological, chemical, medical, engineering or computer sciences) from hypothesis to experiment, observation and evaluation, followed by logical conclusions."

Section 73B(2B)(ii) is further qualified by s 73B(2C) which excludes certain activities from being "systematic, investigative and experimental activities".[37]

Functions of the Board in relation to income tax concessions

132. With effect from 1 October 2001, s 39AA was added to the IRD Act.[38] Section 39AA(1) provides:

"The object of this Part [IIIA] is to complement the tax incentive provided by sections 73B, 73BA, 73I and 73Y of the Income Tax Assessment Act 1936 by giving the Board the role to determine whether eligible companies satisfy the requirements for the incentive."

Section 39AA(2) sets out the object of the sections of ITAA36 specified in s 39AA(1).

An eligible company must be registered under the IRD Act before a deduction is allowable under ITAA36

133. We are concerned only with an eligible company and not with two or more eligible companies jointly registered and so will refer only to the provisions relating to such a company. It is clear from s 73B(10) of the ITAA36 that, without registration of an eligible corporation under s 39J of the IRD Act, a deduction is not allowable under the ITAA36. This is the effect of s 73B(10), which provides, in so far as it applies to an eligible company:

"A deduction is not allowable under this section to an eligible company for a year of income in respect of expenditure in relation to research and development activities unless:

(a) the company is registered, in relation to the year of income and in relation to those activities, under section 39J of the Industry Research and Development Act 1986; or

There then follow detailed provisions relating to the manner in which the amount of the deduction is calculated. They are not relevant.

134. Registration of an eligible corporation takes place under the IRD Act and not under ITAA36. Section 39J in Part IIIA of the IRD Act is the relevant provision. Section 39J(1), with which we are concerned, has remained unaltered over the relevant years. It provides:

(a) an eligible company applies to the Board for registration in relation to activities of either or both of the following kinds in respect of a year of income:

(i) the eligible company's research and development activities;

(aa) the application is in accordance with section 39JD; and

(b) the company provides to the Board such information in relation to the activities covered by the application as the Board reasonably requires;

the Board shall register the company, in relation to those research and development activities, in respect of that year of income."

135. Section 39J is qualified by s 39K. In the relevant years, its sub-sections deal with three discrete subjects. The first was found in s 39K(1) and was concerned with "a ground on which the Board is entitled to refuse" registration:

"Subject to subsection (1A), it is a ground on which the Board is entitled to refuse to register an eligible company, in relation to particular research and development activities in respect of a year of income, that the activities are not research and development activities."

136. The qualification to s 39K(1) is found in s 39K(1A) and relates to advance registration. It provides:

"If advance registration has been granted to an eligible company, in relation to particular research and development activities in respect of a year of income, the Board must not refuse to register the company in relation to those activities in respect of that year on the ground that they are not research and development activities."[39]

137. The second qualification to s 39J is found in s 39K(2). It relates to regulations that may specify other grounds on which the Board is entitled to refuse to register a company in respect of any year of income or in respect of a particular year, or years, of income.[40] Those grounds must be related to matters within the functions of the Board to determine.[41]

138. The fourth qualification set out in s 39K is not so much a qualification to s 39J as the grant of a permission, as it were, to the Board that it need not conduct an exhaustive search for any grounds on which it might exist for refusing registration. Section 39K(4) is framed in this way:

"The Board need not, before registering a company, satisfy itself that no grounds exist for refusing the registration."

It complements s 39J(4), which provides that the registration of a company in respect of a year of income is not invalid merely because a ground existed on which the Board was entitled to refuse to register the company in respect of that year of income.[42]

139. Section 39J(5) provides that "The registration of a company in respect of a year of income is irrevocable." But s 39J(5A) goes on to provide that:

"The Board may alter the registration of a company in respect of a year of income if:

(a) the alteration will correct a mistake in the registration; and

(b) the mistake occurred because the application for registration contained an error in the information required to be specified, or included, in the application under s 39JD(1)(a), (d) or (e); and

(c) the Board determines that it is appropriate for the alteration to be made."

If the Board alters the registration of a company in respect of a year of income, "... the registration has effect, and is taken always to have had effect, as if it had originally been made as altered."[43]

Certificate of non-compliance under s 39N of the IRD Act

140. Section 39N of the IRD Act is concerned with a company's non-compliance with a request the Board might make of a company for particular information relating to activities carried on by or on behalf of the company, or activities in relation to which the company incurred expenditure. The information sought by the Board must be information it needs for the purpose of performing any of its functions, or the exercise of any of its powers, under Part IIIA of the IRD Act. The Board first makes its request of the company but, if the company does not give that information within a reasonable period, the Board may give it a written notice[44] to provide it within 30 days of receipt of the notice. That is the effect of s 39N(1).

141. If a company were to fail to comply with the written notice, the Board may give the Commissioner a certificate that the company has failed to comply in respect of the activities concerned.[45]

Consequences of a certificate of non-compliance

142. Section 73B(33A) of ITAA36 sets out the consequences of non-compliance with a notice under s 39N of the IRD Act:

"Subject to subsection (33C), if the Board gives to the Commissioner a certificate stating that a company has failed to comply with a notice under section 39N of the Industry Research and Development Act 1986 in respect of particular activities, a deduction is not allowable, and shall be deemed never to have been allowable, under this section in respect of expenditure incurred by that company in respect of those activities."

The Board may revoke its certificate. If it does so, the effect of s 73B(33C) is that the certificate is taken never to have been given at all.

143. The Board's decision to issue a certificate under s 39L is the decision under review in this case. Only ss 39L(1) and (2) are relevant[46] and provide:

"(1) The Board may, and shall if requested in writing by the Commissioner to do so, give to the Commissioner a certificate stating whether particular activities that have been or are being carried on by or on behalf of a person were or are research and development activities.

(2) If the Board issues a certificate to the effect that particular activities were not or are not research and development activities, the Board must give notice in writing to the person concerned stating the reasons for issuing the certificate."

1. Meaning of "research and development activities"

144. The expression "research and development activities" is defined in s 4(1) of the IRD Act but that definition does not apply in matters under Part IIIA of the IRD Act. Sections 39A(2) and (3) provide:

"(2) Subject to subsection (1), expressions used in this Part that are defined by section 73B of the Income Tax Assessment Act 1936 have in this Part, unless the contrary intention appears, the same meanings as in that section.

"(3) Subject to subsections (1) and (2), expressions used in this Part that are defined by the Income Tax Assessment Act 1936 (other than section 73B) have in this Part, unless the contrary intention appears, the same meanings as in that Act."

Section 39A(1) does not define the expression "research and development activities". Consequently, the definition of that term appearing in s 73B of ITAA36 is the meaning that must be applied under Part IIIA of the IRD Act.

1. Relevance of a certificate under s 39L to matters decided under s 73B

145. Section 73B(34) of ITAA36 is the complementary provision in the taxation law to s 39L of the IRD Act. It provides:

"If the Board gives to the Commissioner a certificate stating whether particular activities carried on by or on behalf of a specified eligible company were research and development activities, that certificate is binding on the Commissioner for the purpose of making an assessment of the company's taxable income of any year of income in which those activities were carried on."

146. The relevance of a certificate becomes apparent when it is remembered that registration under s 39J of the IRD Act does not itself entitle a taxpayer to a deduction for expenditure incurred in relation to those research and development activities. Registration is a precondition to entitlement but not determinative of it. That is apparent from the wording of s 73B(10) when it is expressed in terms of a deduction's not being allowable in respect of expenditure in relation to research and development activities unless the company is registered in relation to the year of income and in relation to the activities. A clear link is made between the activities in relation to which the expenditure has been incurred and for which a deduction is claimed and the activities in relation to which a deduction is claimed.

147. The clarity of that issue in s 73B(10) is matched by its lack of clarity in s 73B(34) of ITAA36 and s 39L of the IRD Act. There is no clear link drawn in those provisions between the research and development activities in relation to which a deduction is claimed and in relation to which an eligible company has been registered. Indeed, there is no express link at all. Instead, the link must be implied from the circumstances in which the Commissioner requests the certificate. If he does so in the context of considering an eligible company's claim for deduction under s 73B, it must be implied that the activities in relation to which the certificate is sought are those in relation to which the eligible company is registered under s 39J. The Commissioner would have no other basis on which to exercise his power to make a request and, unless the eligible company were already registered in respect of them under, in this example, s 39J, there would be no point in claiming the deduction.

148. The reason why the link is not expressly made in s 39L and even in s 73B(4) lies in the history of the provisions. Before s 39L was added to the IRD Act, it appeared as s 73B(34) of ITAA36. It drew no link between registered activities and those in relation to which the certificate was requested of the Board because registration was limited to that of the eligible company and did not extend to registration of its research and development activities. At that time, registration by the Board was provided for in ss 73B(11) and (12) and not in the IRD Act. The Commissioner would need to be advised by the Board in relation to particular activities when a claim for deduction was made and the activities specified.

149. The effect of the Income Tax Assessment Amendment (Research and Development) Act 1986[47] was to transfer and modify the provisions relating to registration from ITAA36 to the IRD Act. That meant that provisions such as s 39J appeared in the IRD Act in relation to registration and certificates became the subject of s 39L rather than of s 73B(34) of ITAA36. Despite the changes, registration was not made in relation to particular research and development activities. That was so even though the eligible company was required to give information about its research and development activities and s 39K had been introduced and permitted the Board to refuse registration if the activities that were, or are to be, carried on do not include research and development activities. It was clear that, when the Commissioner received a claim for expenditure in relation to specific activities, he might need the Board's view of whether they were research and development activities.

150. Section 39L was amended in 1989 to become a multi-purpose provision serving the needs of both s 73B and Division 10 of ITAA.[48] That was achieved by substituting the word "person" for "eligible company" as the deductions available under Division 10D were not limited to eligible companies. Division 10D provided for deductions for capital expenditure incurred in relation to certain buildings and their extension, alteration or improvement where the buildings, extensions, alterations or improvements were to be used "... for the purpose of carrying on by or on behalf of that person of research and development activities or was for disposal by that person to another person for use for the purpose of the carrying on by or on behalf of that other person of research and development activities."[49] The definition of "research and development activities" drew in the definition in s 73B.[50] No prior registration was required of a person before claiming a deduction under Division 10D. The Board was something that the Commissioner would clearly want to turn to in determining whether the capital expenditure had been incurred for the purposes of research and development activities.

151. In 1996, s 39J(1) was amended to provide for registration of an eligible company in respect of research and development activities.[51] At the same time, s 73B(10) was amended to provide, in effect, that a deduction was not allowable unless the company was both registered and registered in relation to the activities in respect of which it had incurred expenditure for which it was now claiming a deduction. As s 39L continued to have a dual role between s 73B (and ss 73BA, 73I and 73Y also appearing in Division 3 of Part III of ITAA) and Division 10D, it was not amended to draw the link between the activities in relation to which the certificate was given and the activities in relation to which an eligible company was registered. The link must be implied just as the appropriate link must be drawn when the certificate is requested or given in relation to activities in the context of Division 10D of ITAA36.

ANALYSIS OF THE LEGISLATIVE PROVISIONS AND CASE LAW

152. In the previous section of our reasons, we have set out a number of questions that need to be answered in considering a matter under s 39L. Although we consider their order to be logical, we do not think it is the order that best suits our consideration in this case. We will begin with a consideration of whether the activities are "research and development activities" within the meaning of paragraph (a) of the definition of that term in s 73B(1) rather than with whether the activities have been, or are being, carried on and, if so, whether they have been, or are being carried on, for or on behalf of RACV Sales and, if so, whether they have been, or are being carried on, in accordance with guidelines issued under s 39KA.

The criteria identifying "research and development activities"

153. In order to come within the definition of "research and development activities" in s 73B(1) of ITAA36, two main criteria must be met:

(i) "systematic, investigative and experimental activities"; and

(ii) "involve innovation or high levels of technical risk"; and

(iii) are carried on for the purpose of acquiring new knowledge or creating new or improved materials, products, devices, processes or services;

(i) "other activities carried on for a purpose directly related to the carrying on of activities of the kind referred to in paragraph (a)."

Regard must be had to s 73B(2B) in determining whether the activities involve innovation or high levels of technical risk.

154. The word "activities" is used both in the definition of "research and development activities" in s 73B(1) and in s 39L when it requires that the Board's certificate state whether "particular activities" are research and development activities. We understand that word to mean the things "... that people do, especially for pleasure, interest, exercise, etc ..."[52] or other purposes. The reference in s 39L to "particular" activities means that we must decide whether the "... specific; single; individually known or referred to ..."[53] activities, and so "particular activities", carried on are properly characterised as such. We agree with Deputy President McMahon in Re Fermenter and Distiller and Industry Research and Development Board[54] (Fermenter) that this means that "... it is necessary in order to understand the import of those activities, to deal with the background in which they were carried out."[55]

155. In this case, reference is made to projects and the activities undertaken in relation to those projects. It is the activities, and not the project as such, with which we are concerned in the definition. In the case of an eligible company, that is consistent with the fact that registration occurs under s 39J in relation to research and development activities, and not with a project, when an eligible company applies for registration. It is to be contrasted with registration under s 39P. That is concerned with an application for the joint registration of two or more eligible companies in respect of a year of income "in relation to a proposed project or proposed projects comprising or including research and development activities."[56] Although s 39P requires characterisation of both a project and the activities comprised or included in that project, both ss 39J and the provisions of 39L relating to the Board's certificate are concerned only with the characterisation of activities and not of a project of which they may be a part.

When are activities "systematic, investigative and experimental" activities?

156. At all relevant times, the definition of "research and development activities" has referred to the activities' being "systematic, investigative and experimental activities that involve innovation or high levels of technical risk" (emphasis added). Until 19 December 1996,[57] reference was made in the definition of "research and development activities" to "systematic, investigative or experimental activities that ... involve innovation or technical risk ..." (emphasis added). It was in that form that the definition was considered by the Tribunal in Re Charles IFE Pty Ltd and Industry Research and Development Board[58] (Charles IFE) and its words "considered as alternatives and there ... [being] no warrant when assessing applications to require the alternatives to be combined ...".[59]

157. Now that the provision has been amended and the word "and" takes the place of the word "or", Mr McCarthy submitted that we should consider the words to be a composite expression of the sort considered by Drummond J in Secretary, Department of Social Security v Ekis[60] (Ekis). The expression considered by his Honour in that case was "carries on a business" and he noted that ".. Interpreting a composite phrase by dissecting it into its component words and seeking a meaning for each has, however, long been identified as an inappropriate method of construing such a phrase ...".[61]

158. We are bound by that, of course, but note that there is a difference between the composite phrase considered in Ekis and the phrase we must consider. In Ekis, the meaning could not be gleaned by considering first the meaning of "carries on" and then of "business". They were words depending upon each other to convey the meaning. A "business" was not sufficient and nor was "carries on" to convey the meaning conveyed by "carries on a business".[62] The phrase we must consider is a little different. The words are not mutually dependent in the same way. Rather, three adjectives or descriptors are used to describe the activities: "systematic", "investigative" and "experimental". The use of the conjunction "and" between the second and third signifies that each of the words qualifies the activities i.e. they must be "systematic, investigative and experimental activities" (emphasis added). It is not enough, for example, to be systematic and investigative if the activities are not also experimental. They must meet all three descriptors and one descriptor does not qualify another.

159. Taking this approach, we have begun by setting out the ordinary meanings of the words that would seem relevant:

"systematic": "... 1 making use of, or carried out according to, a clearly worked-out plan or method. 2 methodical ..."[63]

"investigative": "investigate ... verb ... to carry out a thorough, detailed, and often official inquiry into, or examination of, something or someone. ... investigative ... adj. ..."[64]

"experimental" "... 1 consisting of or like an experiment. 2 relating to, or used in, experiments. 3 trying out new styles and techniques. ..."[65]

["experiment" "... 1 trial carried out in order to test a theory, a machine's performance, etc or to discover something unknown. 2 the carrying out of such trials. 3 an attempt at something original. ..."[66]]

160. There is little case law to assist us. The words have been used in other statutory contexts where the meaning that they have been given reflects that context.[67] It seems to us that, in the context of research and development and of the tax incentives it attracts, the word "systematic" should be given its ordinary meaning. This interpretation is strengthened in respect of the period after 1 July 2002 for the effect of 73B(2BA) of ITAA36 and of s 39KA of the IRD Act is that activities must be carried on in accordance with a plan complying with guidelines formulated by the Board. A "plan" also signifies "... a thought-out arrangement or method for doing something. ...".[68]

161. "Investigative" is also a word whose nuances of meaning are fashioned by its context.[69] The authorities do not take us any further in understanding it in the context of the IRD Act. In that context, we understand it to have its ordinary meaning so that the activities must involve a thorough and detailed enquiry or examination of a subject matter be it a subject matter that is tangible or intangible.

162. The activities must also be "experimental". Again, the context is relevant in concluding whether activities are experimental[70] but there is nothing in the context of the IRD Act and the definitions it incorporates from ITAA36 that mean that we should adopt anything other than the ordinary meaning of the word.

163. We have looked also at the qualification in s 73B(2C). It provides that fourteen activities or groups of activities cannot be taken to be "systematic, investigative and experimental activities". Should the meaning given to the expression "systematic, investigative and experimental activities" be qualified in some way by reference to the activities that are excluded? Although not expressed in precisely the terms of a proviso, s 73B(2C) is a proviso to that expression whether it is used in the definition of "research and development activities" or in the qualification in 73B(2B)(b) relating to activities involving high levels of technical risk. The way in which a proviso should be interpreted was considered by Latham CJ in Minister for State for the Army v Dalziel.[71] The National Security (General) Regulations made under the National Security Act 1939 permitted the Minister to make orders providing the basis on which compensation was to be awarded for the acquisition of property. The Minister made an order to which Latham CJ turned:

" The order provides that 'the basis of compensation shall not, unless the Central Hirings Committee otherwise determines, exceed the aggregate' of certain sums 'provided that ... (iii) in assessing compensation, loss of occupation or profits shall not be taken into account'. This provision appears as a proviso. Roper J. has interpreted it as a substantive provision and has given it full effect according to its terms, and not merely as a limitation upon the preceding provisions of the regulation. As a general rule a proviso should not be interpreted as if it were a substantive provision independent of the provisions to which it is a proviso. Speaking generally, a proviso is a provision which is 'dependent on the main enactment' and not an 'independent enacting clause': Cf. R. v. Dibdin .... But though a provision framed as a proviso ought to be drafted and generally should be construed only as such, a consideration of both the main and the subsidiary provisions of an enactment may show that the proviso contains matter which is really 'in substance a fresh enactment, adding to and not merely qualifying that which goes before' (Rhondda Urban District Council v. Taff Vale Railway Co.... ). In the present case it is difficult to give effect to the proviso as to loss of occupation and profits merely as a proviso and I think the learned judge was right in treating it as an independent substantive enactment."[72]

164. Applying these principles to s 73B(2C) and to the definition of "research and development activities" and s 73B(2B)(b), it seems to us that s 73B(2C) is intended to be read as a proviso to or as a limitation upon the range of activities that come within the expression "systematic, investigative and experimental activities" but not as a limitation upon the interpretation that should be given to that expression in the first instance. That accords with its drafting. Activities are "taken not to be systematic, investigative and experimental activities" (emphasis added) but that form of drafting recognises that they might have been regarded as activities of that sort but for their being excluded.

165. What amounts to "systematic, investigative and experimental activities" should first be determined according to its ordinary meaning. Section 73B(2C) should then be applied. If the activities come within s 73B(2C), they cannot be "systematic, investigative and experimental activities" even though they would otherwise come within the ordinary meaning of that expression.

When do systematic, investigative and experimental activities "involve" innovation or high levels of technical risk?

166. Paragraph (a) of the definition of "research and development activities" require that the activities "involve" "innovation or high levels of technical risk". The ordinary meanings of the word "involve" include: "... 1 to require as a necessary part. ... ".[73]

167. In Industry Research Board v Unisys Information Services Australia Ltd[74] (Unisys), the Full Court of the Federal Court touched on when activities involve innovation or high levels of technical risk but did so within the confines of a narrow ground of appeal and its conclusion that the Tribunal's view was not determinative of its final decision in any event.

168. Justice Lindgren considered the meaning of "involve" a little further in Industry Research Board v Coal & Allied Operations Pty Ltd[75] (Coal & Allied). His Honour first set out meanings given to them in The Macquarie Dictionary (Macquarie) and in The New Shorter Oxford English Dictionary. Although their narrower or more limited meanings reflected that appearing in Chambers and set out at [61] above, they each had a more liberal meaning. In the Macquarie, the more liberal was: "To include, contain or comprehend within itself or its scope". Justice Lindgren said that:

"... According to either meaning, an activity, although it must be 'particular', is not disqualified by reason only of the fact that in some respects it does not involve innovation or technical risk.

I think that the more liberal meaning is the one invoked by the definition. This seems to be in accord with the approach taken by the Full Court in Unisys and with the Act's object of encouraging research and development activities. Moreover, the more limited meaning is applicable where the word 'involved' is used in relation to an abstract concept, as in 'construction involves work' or 'power involves corruption'. But even if the more limited meaning is in some way the applicable one, I think the AAT was entitled to find it satisfied because the evidence accepted by the AAT showed that innovation and technical risk characterised Construction throughout."[76]

169. Taking Lindgren J's wider interpretation of "involve" the systematic, investigative and experimental activities must include innovation or high levels of technical risk. His Honour's conclusion was expressed with regard to s 73B as it was enacted before 23 July 1996. Since then, s 73B(2B) has been inserted and the definition of "research and development activities" amended. These amendments seem to us to have brought a change of emphasis to s 73B. We explore them further in the following section of our reasons but, for the moment, note that they appear to have shifted the emphasis from a broader meaning of "To include, contain or comprehend within itself or its scope" to that of "to require as a necessary part. ... ". That does not mean that every part of the activity must be a necessary part of, and so involved in, innovation or high levels of technical risk but, when viewed as a whole, the particular activities under consideration must have as a necessary part of them, and so involve, innovation or high levels of technical risk.

When do systematic, investigative and experimental activities involve "innovation or high levels of technical risk"?

170. We will begin with our understanding of the ordinary meaning of the word "innovation" followed by the consideration of it given by the Tribunal and the courts in the context of the definition of "research and development activities" in s 73B(1) before the qualification was introduced through the addition of s 73B(2A).

The qualification was added with effect from 23 July 1996 and so before the period with which we are concerned.

171. The ordinary meanings of the word "innovation" include: "... 1 something new which is introduced, eg a new idea or method. 2 an act of innovating. ...".[77] In Re North Broken Hill Ltd and Industry Research and Development Board[78] (North Broken Hill), the Tribunal considered the ordinary meanings of the word "innovation":

" In the Macquarie Dictionary (2nd Edition) 'innovation' is said to mean 'something new or different introduced; the act of innovating; introducing of new things or methods'. 'Innovate' is stated to mean 'to bring in something new; make changes in anything established; to bring in (something new) for the first time'. In the Shorter Oxford English Dictionary the meaning of 'innovation' is stated as 'the action of innovating; the introduction of novelties; the alteration of what is established by the introduction of new elements or forms'. The meaning of 'innovate', used intransitively, is given as 'to bring in or introduce novelties; to make changes in something established; to introduce innovations'."[79]

172. After analysing the relevant Explanatory Memorandum and Second Reading Speeches, the Tribunal turned to the Frascati Manual which had been given in evidence. It:

"... contains the definition of 'research and development' used by the OECD, to which reference was made in the second-reading speech in the Senate. At page 28 of the manual the basic criterion for distinguishing research and development from related activities is stated to be 'the presence in R&D of an appreciable element of novelty'. That, no doubt, is what is intended by the phrase 'involved innovation' in s 73B(1)."[80]

173. The definition was only said by the Minister to be "based on" the definition used by the OECD and not that "used by the OECD"[81] as the Tribunal stated. Its decision that the activities had to have an appreciable element of novelty was drawn from the Minister's Second Reading Speech. It was a criterion referred to specifically in the Explanatory Memorandum and the Minister had expressly referred to the more detailed explanation of the definition of "research and development activities" to be found in the Explanatory Memorandum:

" The central definition of research and development means activities of a systematic, investigative or experimental nature, involving innovation or technical risk undertaken for the purpose of acquiring new knowledge or for creating new or improved materials, products, processes or services. It thus encompasses each of the elements of basic research, applied research and experimental development, which are explained in greater detail in the Explanatory Memorandum that will be made available to Honourable Senators."[82]

174. A differently constituted Tribunal took issue with this interpretation in deciding Re Confidential and Industry Research and Development Board[83] (Confidential). It rejected the proposition that innovation should be substantial in order for activities to be regarded as research and development activities. Instead, the word should be given a wider meaning.[84] On appeal, the Full Court of the Federal Court endorsed this interpretation in Unisys and added:

" The Tribunal used the word 'substantial' in its reasons in par 34 and its conclusion. In the first case it was to deny any requirement in the statute for substantiality and in the second to describe the real issue as being whether such a degree was requisite. The word 'substantial' has included in its meanings 'of ample or considerable amount, quantity, size, etc.' (the Macquarie Dictionary_, p 1774). It is similarly defined in the_ New Shorter Oxford English Dictionary_, p 3124. It is apparent the Tribunal properly understood these words. The rejection by it_ [the Tribunal] of the proposition substantiality was required did not carry with it the corollary that what remained offended the de minimis principle or accepted ephemeral or nominal innovation or technical work as satisfying the statutory test.

... The word 'innovation' means 'something new or different introduced' and 'the act of innovating: introducing new things or methods' the (the Macquarie Dictionary ...). See also the New Shorter Oxford Dictionary ...."[85]

175. Returning to the word "innovation" as it appears in paragraph (a) of the definition of "research and development activities", there seems to be nothing in the authorities to this point or in the context of ITAA36 that leads us to conclude that it should be given anything other than its ordinary meaning i.e. something new be it a new method or process, a new thing, or new knowledge. It must have that element of "newness" if it is to be innovative but we do not agree with the statement at [9.55] in the Explanatory Memorandum that it is relevant to consider whether "the activities are likely to result in patentable or other protectable intellectual property."

A.2 Innovation: did the "appreciable element of novelty" criterion vary ordinary meaning?

176. Ms Baker submitted that Unisys can be used as authority for the proposition that the introduction of the expression "appreciable element of novelty" in 1996[86] in the form of s 73B(2B)(a) did not effect any substantive change to the ordinary meaning of the word "innovation". She submitted that the word "novelty" is no more than a synonym for the word "innovation" and referred to the definition of the word adopted by the Full Court. The Full Court had adopted the ordinary meaning of the word "appreciable" in Unisys i.e. "capable of being perceived or estimated; noticeable".[87] It seemed to us that Ms Baker was attempting to draw together its understanding of the word "appreciable" and the Full Court's conclusion that the criteria for innovation will be satisfied if innovation is present to some degree provided it is not to a degree that could be described as de minimis.[88] She went on to submit that we could have regard to the Explanatory Memorandum to the Bill[89] that was later enacted as the Taxation Laws Amendment (Research and Development) Act 2001, Act No. 170 of 2001.

177. We will begin with the Explanatory Memorandum to the later amending legislation. Ms Baker referred us to s 15AB(1)(a) of the Acts Interpretation Act 1901 (AI Act), which provides:

"Subject to subsection (3), in the interpretation of a provision of an Act, if any material not forming part of the Act is capable of assisting in the ascertainment of the meaning of the provision, consideration must be given to that material:

(a) to confirm that the meaning of the provision is the ordinary meaning conveyed by the text of the provision taking into account its context in the Act and the purpose or object underlying the Act; ...

Section 15AB(3)(a) requires regard to be had to the desirability of relying on the ordinary meaning conveyed by the text of a provision taking into account its context in the Act and the purpose or object underlying it.

178. The first matter to decide is whether s 15AB(1)(a) permits regard to be had to an Explanatory Memorandum to a Bill amending an Act to assist in the interpretation of an amendment effected by an earlier amending Act. We have found no previous authority that would permit us to do that. To have regard to later material would run the risk of having regard to material that attempts to reinforce what the policy makers intended the law makers to convey rather than to the words that the lawmakers actually used and what those words convey. There can be a disjunct between intention and outcome and:

"... The words of a Minister must not be substituted for the text of the law. ... It is always possible that through oversight or inadvertence the clear intention of the Parliament fails to be translated into the text of the law. However unfortunate it may be when that happens, the task of the Court remains clear. The function of the Court is to give effect to the will of Parliament as expressed in the law."[90]

179. The same thought was also expressed by Kirby P in Avel Pty Ltd v Attorney-General for New South Wales:[91]

"... But the provision of the Interpretation Act which now permit[s] second-reading speeches to be considered in aid of the construction of legislation (see s 34(2)(f)) does not authorise the courts to ignore the language of that legislation. Still less does it permit a court to substitute its view of the intention expressed by the Minister, derived often from discursive language and from material written by departmental officials, for the more precise language of the Bill as enacted. That Bill remains the authentic voice of Parliament, speaking to the courts and to the community. The facility now available by the terms of the Interpretation Act 1987, s 34, does not alter that fact."[92]

180. There can be changes between earlier amendments and views subsequently expressed about them. To have regard to subsequent views would be to run the risk that there is a mismatch between the words of the provision and the views expressed in the later Explanatory Memorandum. As explained by Kirby P in Avel Pty Ltd v Attorney-General for New South Wales, there can be such a mismatch even when the Explanatory Memorandum was prepared contemporaneously with the amending Bill:

"... It is important to compare the Bill to which the Parliamentary remarks are addressed with the Bill as finally enacted. The remarks may be addressed to quite different, and even contrary, language. The Bill may have been substantially amended after the second-reading speech. ..."[93]

181. Bearing these matters in mind, we have had no regard to the Explanatory Memorandum accompanying the Taxation Laws Amendment (Research and Development) Act 2001, Act No. 170 of 2001. There is a further reason why we have put it aside. We do not think that it is an accurate reflection of the law as it stood at the time it was written. It said, in part:

"1.14 The definition of R&D activities has been changed to ensure such activities now involve both innovation and high levels of technical risk. This change has been made to maintain the integrity of the concession by focussing on R&D which involves innovation and high levels of technical risk.

1.15 Past court interpretations of the R&D legislation have established low thresholds for both the 'innovation' and 'technical risk' criteria, such that support under the concession could be extended beyond its policy intent.

1.16 'Innovation' continues to mean an appreciable element of novelty, where novelty means 'something new or different'. 'High levels of technical risk' continues to mean that activities will not involve high levels of technical risk unless there is uncertainty as to whether the technical or scientific outcome can be achieved, and this uncertainty can only be resolved through a program of systematic, investigative and experimental activities. The change to 'innovation and technical risk' is supported by the Frascati definition of R&D activities, which requires the existence of both innovation and technical risk. Such a change will ensure that the focus of R&D activities is where there is uncertainty of outcome and some original thinking is required to resolve the uncertainty."

182. We have had particular regard to [1.15] and looked to the cases to see whether "Past court interpretations of the R&D legislation have established low thresholds for ... 'innovation' ..., such that support under the concession could be extended beyond its policy intent." The Full Court's judgment in Unisys could not be regarded in that way for both it and the Tribunal before it were concerned with the provisions of s 73B before the 1996 amendments. The Tribunal's decision that was the subject of the appeal had identified the years in question as those years of income ending 31 December 1990 and 1991.[94] At [35], the Tribunal noted that s 73B had been amended with effect after the years in question. It misinterpreted the effect of the amendments and this was noted on appeal by the Full Court, which decided that nothing turned on the mistake.[95]

183. What of other cases? Justice Lindgren took a similar view to that in Unisys of the word "innovation". In Coal & Allied he said:

" In Unisys, a Full Court of this Court accepted (at 559) that the word 'innovation' in s 73B(1)(a) meant: 'something new or different introduced' and 'the act of innovating: introducing of new things or methods' for which it cited The Macquarie Dictionary.

I think that this statement by the Full Court accorded to the word 'innovation' an ordinary and broad meaning rather than a narrower technical one, such as one having an affinity with the meaning with 'innovation' in the law relating to patents of invention." [96]

184. That is understandable because the case was decided on the basis of the law as it stood after its introduction in 1986 and after the 1994 amendment. Therefore, it had no reason to consider the amendment that, from 23 July 1996, "activities" "are not taken to involve innovation" unless they "involve an appreciable element of novelty". The same is true of other cases decided in or after 1996 and to which our attention has been drawn: Charles IFE considered issues arising from activities in 1993 and 1994; Fermenter concerned events up to 1 June 1993; and Re The Applicant and Industry Research and Development Board[97] concerned calendar years up to and including 1995.

185. As the cases considered the interpretation of the law as it stood before 23 July 1996, the statement in the Explanatory Memorandum cannot be understood as any reflection of the meaning of the provision that "activities" "are not taken to involve innovation" unless they "involve an appreciable element of novelty". The statement cannot be regarded as reflecting on the meaning of "innovation". Instead, it can only be taken at its face value. That means only that the courts have established low thresholds for the innovation criterion when the law had not introduced "appreciable element of novelty" criterion. The courts established that on the basis of the ordinary meaning of the words as used by Parliament. The policy makers did not intend those words to be understood in that way but policy and intention cannot displace the meaning of the words used by Parliament.

A.3 Innovation: case decided after 1996 amendment

186. The case of Re HZXD and Innovation Australia[98] (HZXD) was decided in relation to activities that had taken place after 23 July 1996. The activities said to be research and development activities were directed to the development of a biometric system of identification so that a customer at a car park would be identified by reference to some personal feature rather than by means of a card or token. Iris and fingerprint technology had been rejected by the applicants but facial recognition had been selected. Evidence was given to the Tribunal that the development of a biometric algorithm designed to permit facial identification was being investigated. Work also had to be done in connecting the biometric information to the software used in a car park to calculate fees payable and to allow entry and egress from it. Evidence was given that, even if the work involved experimentation, it did not involve innovation or technical risk for standard transmission and internet protocols already existed. As facial recognition is not completely secure, a decision had to be made about the level of risk that was acceptable before programming could continue.

" In any event, the evidence of Professor Wagner is to the effect that, while programming work was carried out, it did not even at the time of the claim commencing in 2002 involve 'innovation' or 'high technical risk'. The Tribunal accepts that there is a distinction between conducting trial and error experiments, as was claimed by the applicant to be undertaken during the course of the project, and innovation. The evidence points to the former being undertaken to determine whether, in a system which can never be risk free, an acceptable level of risk can be established. The experiments determined whether one, two or more photographs of the individual's face needed to be taken to ensure as accurate as possible recognition or, if not, whether greater security could be ensured if facial recognition was combined with other modes of recognition (for example, car registration plates). This goes to the efficiency and practicality of the chosen method, but it does not evidence 'innovation'. That is, in the terms of s 73B (2B)(a) of ITA Act, the Tribunal cannot be satisfied that there is 'an appreciable element of novelty'. ..."[99]

On the evidence it had, and we respectfully suggest that there appears to have been a paucity of it,[100] the Tribunal in HZXD had no need to consider the law and could come to its decision on the evidence. We do have to give the law some thought.

A.4 Innovation: the relevance of extrinsic materials regarding the interpretation of "appreciable element of novelty"

188. Unlike the Second Reading Speech given on the introduction of what was to become the IRD Act in 1986, the Second Reading Speech given in 1996 and introducing the "appreciable element of novelty" qualification to the definition of "research and development activities" did not refer to a more detailed explanation in the Explanatory Memorandum. As is to be expected, the Explanatory Memorandum did address the addition of s 73B(2B). It explained that:

"9.52 The definition of 'research and development activities' will be made more explicit by importing concepts from the Explanatory Memorandum to the Income Tax Assessment Amendment (Research & Development) Act 1986. This will make the definition of 'research and development activities' more effective in two ways:

• firstly, the interpretation of the definition of 'research and development activities' is clearer at first glance, and will aid the policy of self assessment in taxation matters; and
• secondly, it puts Parliament's intent for the meaning of 'research and development activities' into a statutory framework which is easier to interpret and administer.

9.54 The meanings of 'innovation', 'high levels of technical risk' and activities which are not taken to be systematic experimental or investigative for the purposes of the definition of 'research and development activities' are taken directly from the Explanatory Memorandum to the Income Tax Assessment Amendment (Research & Development) Act 1986.

9.55 The explanatory notes to the concept of 'innovation' in the Explanatory Memorandum to the Income Tax Assessment Amendment (Research & Development) Act 1986 also apply in respect of the new provisions, namely that innovation may be assessed according to several criteria, including that:

• the activities are seeking previously undiscovered phenomena, structures or relationships;
• the activities are attempting to apply knowledge or techniques in a new way; or
• the activities are likely to result in patentable or other protectable intellectual property.

9.56 Innovation is to involve an appreciable element of novelty. This means that a fairly large constituent part of the activity must involve novelty."

189. The passage in [9.55] of the Explanatory Memorandum is a re-working of the passage relating to "innovation" in the Explanatory Memorandum of 1996. Before we can have any regard to it, we must decide whether we can do so at all. That takes us to s 15AB of the AI Act. In general terms,[101] we may do so in order to confirm that the meaning of a provision is the ordinary meaning conveyed by the text of that provision taking into account its context within that Act and the underlying purpose or object of the Act.[102] We may also do so in order to determine the meaning of a provision when it is ambiguous or obscure, or when its ordinary meaning, as conveyed by its text and taking into account the purpose or object underlying the Act, leads to a result that is manifestly absurd or is unreasonable.[103]

190. When we look at s 73B(2A) of ITAA36, we can see that, in choosing the words "an ... element of novelty", Parliament has chosen an expression that brings to mind patents law and the law protecting intellectual property generally[104] but the question is whether that law has any relevance.

191. In answering that question, two matters come to mind. The first is patentable inventions[105] are not determined solely by reference to novelty or even innovation. As provided in s 18(1) of the Patents Act 1990 (Patents Act), for example, an invention is a patentable invention for the purposes of either a standard patent or an innovations patent if the invention is, among other matters, novel when compared with the prior art base as it existed before the priority date of the claim. How that is to be determined is the subject of s 7(1).[106] Other criteria that an invention must meet in order to be patentable are that it "is useful" and "was not secretly used in the patent area before the priority date" of the claim by, or on behalf of, or with the authority of, the patentee or nominated person or their predecessors in title to the invention.[107]

192. The second thing to bear in mind is that patents law is concerned with the "sole working or making of any manner of new manufactures" within the meaning of s 6 of the Statute of Monopolies (UK). When used as a noun, the word "manufacture" refers to "... 1 the practice, act or process of manufacturing something. 2 anything manufactured. ...".[108] It is not used to describe the activities that lead to the development of those "manufactures" be they a "thing", whether tangible or intangible or a process of manufacturing a thing. Section 73(2B)(a) of ITAA36 is concerned with the activities that lead to the development of the "manufactures" and not with the "manufactures" themselves.

193. It would seem, then, that it is not appropriate to take the concept of "an ... element of novelty" as it is used in intellectual property law and to apply it to the concept of "an appreciable element of novelty" as it is appears in s 73B(2A)(a). When patents law is taken out of contention, there is nothing in the provision that is ambiguous or obscure.

194. We have looked also at whether the ordinary meaning of s 73B(2A)(a) when taken in its context and having account of the purpose of the statutory regime under s 73B leads to a result that is manifestly absurd or unreasonable. We do not think that it does. The criterion may not be easy to apply in particular factual circumstances but its ordinary meaning seems quite sensible and entirely in accordance with the object expressed in s 73B(1AAA). That leaves s 15AB(1)(a) of the AI Act, which permits reference to be made to extrinsic material to confirm that the ordinary meaning conveyed by the text of a provision when taken in its statutory context and having regard to the object of the Act.

A.5 Innovation: ordinary meaning of "appreciable element of novelty"

195. The phrase "appreciable element of novelty" is a composite phrase and should be interpreted as such but, to understand such a phrase, we must have some understanding of the ordinary meanings of the words that form it. They are:

"appreciable": "... 1 noticeable; significant; able to be measured or noticed ..."[109]

"element": "... 1 a part of anything; a component or feature. ..."[110]

"novelty": "... 1 the quality of being new and intriguing. 2 something new and strange. ..."[111]

196. The word "appreciable" is an adjective whereas the remaining two are both nouns. The adjective could be read as a descriptor of the word "element" but, given that the noun "novelty" is acting as a descriptor of the word "element", we think it better to read the adjective "appreciable" as the descriptor of the compound expression "element of novelty". That means that the element of novelty must first be identified and the question then asked whether it is appreciable.

197. In what is the element of novelty to be identified? It lies in the "activities". That follows from the words of s 73B(2B)(a) when it says that the "activities are not taken to involve innovation unless they involve an appreciable element of novelty" (emphasis added). That means that the activities must first be identified. That is consistent with that part of the definition of "research and development activities" referring to "... activities ... that involve innovation ...". Once identified, the question becomes: Do these activities have a feature or a component that is new (an element of novelty)? If so, the next question becomes: Is that element of novelty something that is significant in some way (appreciable)? The activities must be viewed as a whole and in context in order to answer these questions. That seems to us to be the ordinary meaning of the words in the context of s 73B(2A)(a), s 73B and of the scheme provided in ITAA36 and IRD Act.[112]

198. If we understand the submission correctly, Ms Baker submitted that s 73B(2B)(a) does not require any original thinking or new ideas to be introduced into the process. It is enough that there is new knowledge obtained and introduced through the activities that have been undertaken or new products such as, in this case, meaningful crash test ratings. She relied on the judgment of the Full Court of the Federal Court in Unisys when it set out a passage from the reasons for decision of the Tribunal. The Full Court had set out [37] from the Tribunal's reasons. It read:

" The respondent submitted that merely to produce a new product, or one having novelty, does not necessarily involve innovation under the Act. As counsel put it 'it is the process, not the product, that must meet the criteria, as Re North Broken Hill Limited shows'. In our opinion, this proposition cannot be elicited from anything that was said in the reasons in Re North Broken Hill Limited. The legislation requires that the activities involve innovation. If what is produced is an innovative product then the activity must, as a matter of logic, involve innovation. The definition in the ITAA in fact contemplates that the purpose of the research may be the creation of new or improved products. Neither the definition, nor anything said in the previous cases require that this reference be read down to refer only to processes rather than products."[113] (emphasis added)

199. Miss Baker drew particular attention to the last sentence in [37] and to the same point subsequently made by the Tribunal in Re The Applicant and Industry Research and Development Board:

"_The statement that it is the process not the product that must meet the criteria is not correct. There is nothing in the legislation requiring that the definition be read down to refer only to processes rather than products (_Unisys at first instance not dissented from by the Full Court)."[114]

200. We would respectfully suggest that there is a considerable difference between the Full Court's not dissenting from a proposition and its agreeing with a proposition. The considerable difference lies in the fact that the Full Court might not have turned its mind to the issue at all. That is not because it did not carry out its role but because it did. Its role in Unisys was to hear an appeal lodged under s 44 of the Administrative Appeals Tribunal Act 1975 (AAT Act). Section 44(1) provides that "A party to a proceeding before the Tribunal may appeal to the Federal Court on a question of law, from any decision in that proceeding." There has been much judicial discussion of what is meant by a "question of law" but we do not need to consider it here. It is enough to set out the general statement of law by in Clements v Independent Indigenous Advisory Committee[115] by Gyles J who, though in dissent, was not in dissent on general principles:

"61 Section 44 of the Act refers to a question of law, not an error of law. Section 45 of the Act casts some light upon the meaning of that phrase when it permits the Tribunal to refer 'a question of law arising in a proceedings' to the Federal Court for decision. This, to my mind, assists in coming to the conclusion that 'question of law' in s 44 refers to a question of law which may be posed as such rather than supervision of the way the Tribunal carries out a review. This is consistent with the general legislative scheme. Questions of natural justice more sensibly fall within the purview of the ADJR Act than the special and limited provisions of s 44.

62 This view is reinforced by the constitutional backdrop against which the legislation is to be viewed. The Administrative Appeals Tribunal is a purely administrative body, and does not exercise any of the judicial power of the Commonwealth. The Federal Court can only exercise the judicial power of the Commonwealth. Answering questions of law which have arisen in the course of the proceedings of the Tribunal can be seen to be the exercise of that power. Supervising the procedures of an administrative body as such is not the exercise of the judicial power of the Commonwealth. It is reasonable to assume that the legislature would have been anxious not to blur the role assigned to the Federal Court by s 44. Any necessary judicial review could be undertaken pursuant to the ADJR Act (cf Minister for Immigration & Ethnic Affairs v Gungor [1982] FCA 99; (1982) 42 ALR 209 per Fox J at 211-212 and Sheppard J at 218-223; TNT Skypak International (Aust) Pty Ltd v Federal Commissioner of Taxation [1988] FCA 119; (1988) 82 ALR 175 particularly at 181)."[116]

201. It follows that if, on an appeal under s 44 of the AAT Act, there is no question of law raised by the appeal, the Federal Court does not have jurisdiction to decide the issue.[117] In Unisys, the Full Court did not set out the precise grounds of appeal but referred to them more generally. After summarising the Tribunal's reasons, it dealt with five grounds of appeal directed to the manner in which the Tribunal had interpreted the words "innovation", "technical risk" and "innovation or technical risk". It then turned to the statement made by the Tribunal in [37] that "... if what is produced is an innovative product then the activity must, as a matter of logic, involve innovation ...".[118] The Full Court noted that this sentence had been preceded by the statement that "The legislation requires that activities involve innovation."[119] The Full Court dealt with the matter in this way:

" Even if it is the case an innovative product may result from an activity which does not involve innovation (a proposition which should be the subject of more extensive argument than occurred before us), we do not consider the Tribunal's view on this issue was determinative of its decision in the matter under appeal. The conclusion the Tribunal reached was one relating to the activity and to whether it possessed the character of involving innovation or technical risk. It did not arrive at its conclusion by inference from the nature of the product. The issue arose before it only in answer to a submission made to it by counsel for the applicant. There was no relevant error of law in the Tribunal's statement."[120]

202. As we understand this passage it is not an endorsement of the Tribunal's statement in Confidential that "Neither the definition, nor anything said in the previous cases require that this reference be read down to refer only to processes rather than products".[121] What their Honours were saying was:

(2) it would have to be fully argued if we were to express a view;

(3) it is not relevant for us to express a view because:

(a) the Tribunal decided the matter by reference to activities and not by reference to the product either expressly or by inference from the nature of the product; and

(b) the Tribunal only considered the matter at all because counsel raised it; not because it was relevant to do so; and

(4) "There was no relevant error of law in the Tribunal's statement." (emphasis added):

(a) meaning that there was no error of law that affected the particular decision made by the Tribunal; but

(b) not meaning that there was necessarily no error of law in the Tribunal's general statement of the law that was a statement going beyond the basis upon which the Tribunal had reached its decision.

203. Ms Baker also drew our attention to the judgment of Lindgren J in Coal Allied. Again, the relevant passage at [26(i)] of the Tribunal's reasons in Re The Applicant and Industry Research and Development Board were set out by his Honour in Coal & Allied.[122] His Honour did not return to the point and his judgment centres on the activities that were undertaken and whether they came within the definition of "research and development activities". He drew a distinction between a project (which he said was the wrong focus) and activities (which he said were the correct focus). In those circumstances, his judgment, we respectfully suggest, cannot be taken as support for the proposition that the definition of "research and development activities" may be met by reference to products as much as processes.

204. In any event, the proposition put forward by Deputy President McMahon in Confidential cannot be accepted. The definition itself distinguishes between activities and their outcomes be they described as products, knowledge, devices and so on. As a matter of grammar, regard cannot be had to the outcomes until the activities have been identified, found to be of the sort described in the opening lines of the definition and being carried on for the purposes, and so to achieve the outcomes, in paragraphs (a)(i) and (ii). The amendment to change the description of the activities from "systematic, investigative or experimental", as was the case when Deputy President McMahon considered the definition to "systematic, investigative and experimental", as it reads now, makes no difference to our view on this aspect.

205. Quite apart from the terms of the definition, we do not understand how such a proposition can stand. The further proposition put forward by Deputy President McMahon in Confidential and preceding his statement that nothing requires reference to be made only to processes rather than products was this: "If what is produced is an innovative product then the activity must, as a matter of logic, involve innovation."[123] He then makes the point that the definition contemplates that the purpose of the research may be the creation of new or improved products.

206. We do not agree with Deputy President McMahon's logic. A carpenter might make an innovative piece of furniture. Why does that fact lead to the conclusion that the carpenter did so using innovative methods of construction? Why could the carpenter not have made it using traditional methods? Without more, either proposition is as likely to have been the case as the proposition that the carpenter used some innovative methods and some tradition methods. As Kitto J said in Jones v Dunkel:[124]

"... One does not pass from the realm of conjecture into the realm of inference until some fact is found which positively suggests, that is to say provides a reason, special to the particular case under consideration, for thinking it likely that in that actual case a specific event happened or a specific state of affairs existed. ..."[125]

207. Before its amendment in 1996, the definition of "research and development activities" required, among other things, that the activities involve "technical risk" if they did not involve innovation. Since 23 July 1996, they must involve either innovation or "high levels of technical risk" (emphasis added). Despite the change, it remains relevant to look to the previous authorities to see their interpretation of the words "technical risk" before attempting to reach an understanding of the criterion as it is currently drafted.

208. In Re Mobil Oil Australia Limited and Industry Research and Development Board[126] (Mobil Oil Australia), the Tribunal adopted the meaning that appeared to it to have been agreed upon between the parties. It had been expressed by Mr Forsyth QC, who had appeared with Mr Murphy of counsel for the applicant, in this way:

"As far as technical risk is concerned, the dictionary is not very helpful here because of the combination of the two words but there seem to be general agreement amongst the experts that it is referring to uncertainty as to outcome and that indeed seems to be the logical meaning of the words in the context. So there has to be a relevant uncertainty as to outcome. And we would accept once again that it has to be a material uncertainty. As my learned friend says there is always some element of uncertainty at the margin, complete precision is not possible in whatever you do and we would agree that very minor uncertainty as to outcome certainly will not qualify, it has to be a material uncertainty but once again that is not a vigorous hurdle."[127]

We are loathe to adopt this interpretation without further analysis for there is no finding in the Tribunal's reasons that the expression "technical risk" has a technical or trade meaning. Only if it has, may evidence be received of its meaning[128] and then only in accordance with the principles set out by Hill J in Pepsi Seven-Up Bottlers Pty Ltd v Federal Commissioner of Taxation.[129]

209. In Re The Applicant and Industry Research and Development Board,[130] Deputy President McMahon said:

"(e) The meaning of 'risk' is 'uncertainty as to outcome' and the word 'technical' qualifies 'risk' adjectivally and means 'belonging or pertaining to an art, science or the like'. An activity may, therefore, be innovative because the implementation had to pioneer new territory and may involve technical risk because, while there was little risk in the technology itself, there was induced risk in the attempt to apply it (Mobil Oil Australia)."[131]

210. We do not understand Mobil Oil Australia to be authority for the latter part of this proposition. The Tribunal set out the evidence and summarised the submission on the point made by senior counsel for the applicant.[132] Following its reference to the meaning of "technical risk", its only reference to the subject was:

"The applicant contended that the key issues involved in processing a mixture of Arab and Gippsland crudes at Altona were sulphur emissions and the pour point and sulphur content of certain products.

We find that the refinery trial did not involve technical risk."[134]

211. On appeal from the Tribunal's decision, the focus of the Full Court's consideration in Unisys was more on its conclusion that something more than de minimis compliance with the criteria was required. Three relevant passages from the judgment are:

" The word 'risk' is defined to mean 'exposure to chance of injury or loss; a hazard or dangerous chance' ... (the Macquarie Dictionary_, p 156). Se also the_ New Shorter English Dictionary_, p 2609_."[135]

"In our opinion there is nothing in the text of the Act which put into question the meaning of the words 'innovation or technical risk' and required confirmation the words were being used in their ordinary meaning. There is nothing in the words or the Act which indicates the words were intended to have any meaning other than their respective ordinary meanings. See: Collector of Customs v Agfa-Gevaert Ltd [1996] HCA 36; (1996) 141 ALR 59 at 62-65."[136]

"... It is clear that, provided the condition of innovation or technical risk is not de minimis, the conditions will be satisfied by the presence of innovation or technical risk of whatever degree and not necessarily of any particular degree. One activity may be considerably more innovative and possessed of risk than another but each may satisfy the conditions of the statutory language."[137]

212. In Coal & Allied, the Board submitted that the Tribunal had understood "technical risk" to mean "mere uncertainty as to outcome" whereas what was required was a realistic risk of failure. In concluding that the Tribunal had not erred in its understanding of what amounted to "technical risk", Lindgren J did not address the matter in terms of degree of uncertainty of outcome or realistic or otherwise risks of failure. Instead, he canvassed the evidence and concluded:

" In my respectful opinion, what I have said above and earlier about Mr Davidson's eight technical risks shows that there was ample evidence before the AAT that the Construction activity was attended by risks and that those risks were of a technical nature. It is not the point that it was possible to conceive at the Design stage of an area of potential difficulty in the later Construction stage. There remained uncertainty as to whether what was to be tried at first would succeed. If it did not as in the case of the attempt to excavate coal by the use of a backhoe or clamshell, there was the risk of delay and extra cost while an alternative was found. No doubt the extent of exposure to potential loss varied as between risks."[138]

213. If the activities are not innovative, they must involve "high levels of technical risk" and must meet the criteria in s 73B(2B)(b). Before 23 July 1996, they needed only to involve "technical risk" and we will start our consideration with that expression. It is the expression considered in earlier authorities and, although not determinative of our consideration, sets the background for it.

214. The ordinary meanings of each the words in the expression "technical risk" do not assist us greatly. Those for the word "risk", when used as a noun as it is here, include "... the chance or possibility of suffering loss, injury, damage, etc; danger ...".[139] The word is qualified by the word "technical". The ordinary meanings of that word include:

"... possessing knowledge of, specializing in or relating to practical skill or applied science, especially those sciences which are useful to industry ..."[140]

A better understanding of the word "technical" comes from an understanding of the way in which the word is used to qualify other nouns:

"technical college ... a college of further education that teaches practical skills and applied sciences that are necessary for industry and business ..."

"technical drawing ... 1 the drawing of plans, machinery, electrical circuits, etc done with compasses and rulers, etc for business and industry. 2 a drawing done for business and industry."

"technical hitch ... 1 a mechanical fault which causes a temporary halt to proceedings. 2 loosely a snag or problem."[141]

215. When regard is had to the way in which the word "technical" is generally used to qualify a noun, it would seem to us that the ordinary meaning of the expression "technical risk" is that there is a chance or possibility that the processes or means (be they classified as mechanical, electrical, scientific, technological or in some other way that can be described as "technical") involved in the activities will fail to work, to play the part intended for them or to achieve an outcome intended for them. How failure is described and what amounts to failure will depend upon the particular technical discipline concerned. Whether there is a chance or possibility of failure depends upon evidence relevant to that discipline.

216. Failure of this sort may lead to the outcome that the activities will fail to achieve their purpose but not necessarily so. Technical processes or means may themselves fail and yet, for reasons not necessarily foreseen, achieve their purpose. Achievement of the purpose for which the activities are carried out is not a relevant consideration in assessing technical risk. Whether in its earlier or current form, the definition of "research and development activities" describes first the activities and then the purpose for which they are carried on. That they contain technical risk or high levels of technical risk is as much a description of the activities as that they are systematic, investigative and/or experimental. The description of the activities and their purpose are quite separate and one does not qualify the other.

B.3 Technical risk: "high levels of technical risk"

217. Since 23 July 1996, it has not been enough that there is technical risk. There must be "high levels of technical risk". Disregarding for the moment any effect of the qualification in s 73B(2B)(b), that means that a mere chance or possibility of technical failure is not enough to justify a finding that activities involve technical risk. The ordinary meanings of the adjective "high" include "... intense or advanced; more forceful than normal ..."[142] and "... intensified; exceeding the common degree or measure; ...".[143] The reference to a "level" is a reference to the "degree"[144] in the context of risk. That means that, if activities involve high levels of technical risk, there must be a significantly greater chance or possibility that the technical processes or means involved in the activities will fail than would normally be the case. What would normally be the case will be ascertained by reference to the industry or field of endeavour in which the activities are being conducted. It will be a matter of evidence. It, and so the level of risk, cannot be measured by reference to legal standards such as whether it is "more likely than not" that there will be technical failure.

B.4 Technical risk: the qualification in s 73B(2B)(b)

218. This is the interpretation of "high levels of technical risk" as it appears in the definition of "research and development activities" but it is not the end of the matter. Regard must be had to s 73B(2B)(b). It narrows the range of activities that may be regarded as involving high levels of technical risk. It does this in two ways.

219. Section 73B(2B)(b)(i) does this not by reference to chances and possibilities of failure of technical processes and means but by reference to an inability to know or determine, in advance, the probability of obtaining a technical or scientific outcome on the basis of current knowledge or experience. They are really two sides of the same coin with one expressed in terms of determining the likelihood of failure of achieving an outcome and the other in terms of obtaining an outcome.

220. The inability to know or determine the probability of obtaining a technical or scientific outcome is not determined by reference to the knowledge or experience of those undertaking the activities. Section 73B(2B)(i) is expressed in the passive voice and so on objective terms. Reference must be made to the "knowledge or experience" that is available generally before it can be decided that the outcome of obtaining a technical or scientific outcome cannot be known or determined in advance.

221. It is not enough that the outcome cannot be known or determined in advance. Section 73B(2B)(b)(ii) is addressed to the criterion in s 73B(2B)(b)(i) that the outcome cannot be known or determined. It effectively defines the only means by which uncertainty of obtaining an outcome can be removed. It requires that the uncertainty of obtaining the outcome can only be removed through a program of activities that must meet the description it sets out. That means that the not knowing or not being able to determine, and so the uncertainty,[145] of obtaining the technical or scientific outcome can be removed only through that program of activities. Removing that uncertainty does not mean ensuring that a particular technical or scientific outcome is assured. It has nothing to do with a particular outcome and everything to do with being able to assess whether a particular technical or scientific outcome can be obtained at all.

222. The activities to which s 73B(2B)(b)(ii) refers are not necessarily the activities referred to in paragraph (a) of the definition of "research and development activities" in s 73B(1). They may be but it is not necessary to decide the point. Even if they do not, they would fall under paragraph (b) as activities carried on for a purpose directly related to carrying on those in paragraph (a).

223. We have referred to the requirement that the uncertainty of obtaining an outcome can be removed only through activities meeting the description in s 73B(2B)(b)(ii). We have already considered the meaning of the expression "systematic, investigative and experimental activities".[146] There must be a "program" of such activities. That means that the activities must be comprised within a "plan or schedule"[147] of such activities.

224. "Scientific method" must be applied in that program of activities in a "systematic progression of work ...[as described] from hypothesis to experiment, observation and evaluation, followed by logical conclusions." A "scientific method" would suggest that the work is done according to "an ordered set of procedures or an orderly system. ..."[148] as it would be in science and "... displaying the kind of principled approach characteristic of science".[149] The reference to science does not limit the method to any particular activity for it extends to "... any area of knowledge obtained using, or arranged according to, formal principles. ...".[150]

225. The work must be based on principles of physical, biological, chemical, medical, engineering or computer sciences. The orderly way in which the work must be conducted is emphasised by the stipulation that the work take a certain course. It must begin with an hypothesis i.e. "... 1 a statement or proposition assumed to be true for the sake of argument. 2 a statement or theory to be proved or disproved by reference to facts. 3 a provisional explanation of anything."[151] Given the context of research and development, it would seem to be an implicit requirement that the hypothesis be framed so that it is relevant in, or directed to, removing the uncertainty of obtaining the technical or scientific outcome of the activities. That means that it is relevant in, or directed to, determining the probability of obtaining the technical or scientific outcome of the activities.

226. There must be a "progression" of that work so that it can be said to be "... moving forwards or advancing in stages ..."[152] to "experiment, observation and evaluation". Given the context of research and development, the experiment, observation and evaluation must be directed to testing the hypothesis. What work amounts to experiment, observation and evaluation will depend very much on the particular discipline in which it is being undertaken. That is a matter for evidence. Having done so, the work must move to drawing logical conclusions from the experimental, observational and evaluative work. Those logical conclusions must relate back to the hypothesis.

"activities ... carried on for the purpose of" acquiring new knowledge or creating new or improved materials etc.

227. A "purpose" in the context of the definition in s 73B(1) of ITAA36 is "... the object or aim in doing something. ... the function for which something is intended ...".[153] It is not a rationale developed at a later time to explain why, in this case, activities were carried on at an earlier time. It must be the object or aim in undertaking or continuing to undertake the activities and so the object or aim for which those activities were "carried on".[154] It may be arguable that the definition of "research and development activities" does not require it to be the sole purpose in undertaking the activities but that is not a matter we need decide.

228. The purpose is expressed in terms of the acquisition of "new" knowledge and the creation of "new or improved" materials, products, devices, processes or services. The word "new" is used in the sense of "recently discovered ... [or] never having existed before; just invented, etc. ...".[155] That is the ordinary meaning of the word. We do not think that there is anything in the definition of "research and development activities" or elsewhere in s 73B that suggests that additional criteria relevant in determining what is patentable should be read into the definition whether reference is made to knowledge or materials. The reference to "creating ... improved materials" should also be understood according to its ordinary meaning. That is of forming or producing materials that are "... better or of a higher quality or value ..."[156] than materials of that type already in existence.

229. This interpretation is consistent with the purpose of the scheme when it was introduced in 1986. At that time, it was to take its place as "... part of a broader set of objectives which seek to encourage, through the Governments' industry and technology policies, the development in Australia of internationally competitive, export oriented innovative industries."[157] The particular objects included those to provide incentives for greater research and development in Australia, to support it, and to exploit technological developments occurring overseas. It may well be that industry and others would want to protect their investment by means of the intellectual property law but Parliament has not made that, or the likelihood of their doing so, a criterion for recognition of activities as involving innovation. There is, however, nothing in the words that Parliament has used to suggest that it is relevant to consider whether "the activities are likely to result in patentable or other protectable intellectual property." Therefore, we do not agree with the statement to that effect at [9.55] in the Explanatory Memorandum to the Bill that was to become the Taxation Laws Amendment Act (No 3) Act 1996.

"other activities that are carried on for a purpose directly related to the carrying on of activities of the kind referred to in paragraph (a)"

230. Justice Lindgren considered paragraph (b) of the definition of "research and development activities" in Coal & Allied. Other than to omit a requirement regarding the location of the activities, the substance of the paragraph has not changed. We think that we are bound to adopt his Honour's interpretation of the paragraph when he said:

"... I agree with the AAT that the use of the article 'a' in par (b) indicates that the purpose specified in the paragraph need not be the 'sole' or 'dominant' or 'actuating' purpose, and that it is sufficient that it was simply 'a' purpose of the activity in question."[158]

231. It must, though, be a purpose "directly related to" the carrying on of activities of the kind referred to in paragraph (a) of the definition of "research and development activities". As Lindgren J said in Coal & Allied:

"... One activity can be carried on 'for a purpose directly related to' the carrying on of another activity even though that other activity is completed by the time the activity in question begins."[159]

232. Whether one activity is "directly related to" another suggests that relationship between the two is immediate but the words are dependent upon their context. When considering whether mesh screens opened "directly to the outside" Basten JA said:

"... In ordinary usage, the word 'directly' is to be contrasted with 'indirectly'. It may well have different connotations in different contexts: compare the Constitution, ss 7 and 24 requiring that members of Parliament be 'directly chosen by the people'; Insurance Commission (WA) v Container Handlers Pty Ltd [2004] HCA 24; (2004) 218 CLR 89, discussing 'injury ... directly caused by, or by the driving of, the vehicle'. In relation to an opening in a wall, the term suggests a lack of space between the place in question and 'the outside'. The context does not permit a precise meaning for the phrase taken as a whole. Clause 6, in common with s 23 of the Act, appears to distinguish between 'an enclosed public place' and 'a covered outside area': cl 6(1). In relation to a public place within a building, it may be, as the magistrate appears to have assumed, that 'the outside' means outside the building. Even if the outside is beyond the limits of a building, there may well be, as the magistrate explained, practical considerations, such as those created by an opening onto a verandah or an area under eaves."[160]

233. Like an opening in a wall, whether an activity is directly related to another will be a matter to be decided in a particular case. It will be relevant to consider whether there are other factors that need come into play or steps that need to be taken before it can be said that one activity is carried on for a purpose related to another. If there is such a need, it may be one is not directly related to the other.

234. The requirement that the activities be "of the kind referred to in paragraph (a)" must be a reference to the activities being "of the same sort"[161] as those referred to in paragraph (a). That means that there will need to be an analysis of the activities to which they are said to be directly related. That analysis may have already been undertaken because a taxpayer may seek a deduction in respect of activities claimed to come within both paragraphs of the definition of "research and development activities" as understood by reference to s 73B(2B). It may not have been undertaken if a taxpayer claimed a deduction only on the basis of activities said to come only within paragraph (b) of the definition of "research and development activities". That might arise if, for example, the taxpayer were not carrying on the activities of the kind referred to in paragraph (a).

235. Section 39T(1) of the IRD Act provides that an application may be made to the Tribunal for review of certain decisions made by the Board. There is no question that the decision of which RACV Sales seeks review is such a decision. Section 39T(4) provides:

"The hearing of a proceeding relating to a decision covered by subsection (1) must take place in private and the Administrative Appeals Tribunal may, by order:

(1) give directions as to the persons who may be present; and

(b) give directions of a kind referred to in paragraph 35(2)(b) or (c) of the Administrative Appeals Tribunal Act 1975."

In the absence of an order under s 35 of the AAT Act, we do not consider that this section either justifies or permits an applicant to be identified by a pseudonym. We will explain why we hold that view.

236. The word themselves suggest that a private hearing is a hearing that is "not open to ... the general public. ...".[162] As Lockhart J described it in National Companies and Securities Commission v Bankers Trust Australia Ltd:[163]

"... The essence of a private hearing before the Commission is that what takes place is in private and, therefore, by definition and of necessity not open or accessible to the public."[164]

What restrictions are consequent upon a private hearing?

237. The mere fact that a hearing is not open to the public does not automatically lead to the conclusion that what is said and done at that hearing cannot be spoken of publicly. This was addressed by the House of Lords in Scott v Scott.[165] That was a case in which the majority of the House of Lords decided that the Judge at first instance had not had jurisdiction to order that a nullity suit in the matrimonial causes jurisdiction be heard "in camera" and so in his chambers (room or office) at the court. Therefore, neither Mrs Scott nor her solicitor had been in contempt of court when they sent copies of the transcript of the proceedings to certain persons in order to defend her reputation.

238. Despite that conclusion, their Lordships went on to consider when a matter should be heard in private and whether what transpired in private could be made known. In this case, the former issue is not relevant for it has been determined by Parliament that the hearing will take place in private. The latter issue is relevant. In summary, their views were:

"... While the broad principle is that the Courts of this country must, as between parties, administer justice in public, this principle is subject to apparent exceptions ... As the paramount object must always be to do justice, the general rule as to publicity, after all only the means to an end, must accordingly yield. But the burden lies on those seeking to displace its application in the particular case to make out that the ordinary rule must as of necessity be superseded by this paramount consideration. The question is by no means one which, consistently with the spirit of our jurisprudence, can be dealt with by the judge as resting in his mere discretion as to what is expedient. The latter must treat it as one of principle, and as turning, not on convenience, but on necessity."[166]

The Earl of Halsbury agreed generally with Viscount Haldane but added that "... as to the injunction of perpetual secrecy, there is not a judgment of authority to justify it. ..."[167]

Lord Loreburn decided that s 46 of the Matrimonial Causes Act 1857 permitted the trial judge to order a private hearing. It permitted rules to be made regulating hearings otherwise than in open court but no such rules had been made. As they had not "... the Divorce Court is bound by the general rule of publicity applicable to the High Court and subject to the same exception. ..."[168]

"... what on its face it plainly says, and nothing more, namely, this, that the place where the case is to be heard shall be a private chamber, not a public Court. ... The order was, I think, spent when the case terminated, and had no further operation beyond that date. ..."[169]

" I am of the opinion that the order to hear this case in chambers was beyond the power of the judge to pronounce. I am further of the opinion that, even on the assumption that such an order had been within his power, it was beyond his power to impose a suppression of all reports and what passed at the trial after the trial had come to an end. ..."[170]

239. Scott v Scott has been considered in a number of cases in Australia. In Attorney-General (NSW) v Mayas Pty Ltd,[171] Mahoney JA summarised:

" There are two basic principles affecting proceedings in courts. Stated broadly they are: that the hearing of proceedings in a court - and, I think, of criminal proceedings in particular - should be open to the public; and that what takes place in those proceedings may be the subject of a fair and accurate report and of appropriate discussion. These two principles, while they are related, are distinct, in their operation and their rationale.

These principles are, of course, not absolute. They are subject to qualifications. In particular, they are subject to, as I shall describe it, the Scott qualification ... Over-simplified, the Scott qualification authorises the court to depart from these principles if it is necessary to do so in order that justice be done in accordance with law.

The Scott qualification, that is, the considerations on which it is based, can authorise both in camera and non-publication orders. Proceedings may be ordered to be held in private and some or all of the public excluded: this is what was in issue in the McPherson case. And it may prevent the publication of what took place: this was what was in issue in the Scott case. A nonpublication

order may be made for the purpose of making effective the exclusion of the public: this was what was discussed, for example, by Earl Loreburn in the Scott case. But it may also be made for other reasons. Thus, where proceedings are not held in camera, a non-publication order may be made to protect an informer ...Non-publication may, I think, be ordered in blackmail cases:... Non-publication orders may, in my opinion, be made where necessary in the interests of national security. ...

As I have indicated, sometimes the making of a non-publication order is justified because otherwise an in camera order would be ineffective and the purposes for which it was made would not be achieved. It is, I think, obvious that in some cases this may be so. The extent of the restrictions imposed by an in camera order, as such, is not free from doubt. It might be thought that, at least in some cases, the reason why an in camera order can validly be made should lead the court to conclude that, by implication, the in camera order also restricts publication. But in England it has been said that an in camera order does not, as such, restrict the subsequent publication of what has taken place in the court: see the Scott case (at 483) and Halsbury's Laws of England, 4th ed, vol 9, par 20(10) at 17. If this be so, then it would follow that in some but not all cases in which it is appropriate that an in camera order be made, the considerations which made the in camera order appropriate would justify the making of a non-publication order: see, eg, the Scott case (at 447, 451)."[172]

240. These principles were considered by the Full Court of the Federal Court in National Companies and Securities Commission v Bankers Trust Australia Ltd. Although Lockhart J dissented regarding the terms of the order that had been made, the Full Court agreed that s 36(2) of the National Companies and Securities Commission Act 1979 conferred, by necessary implication in its context, power upon the Commission to take all reasonable steps, at least during the hearing itself, that were necessary to ensure that the hearing was conducted privately.[173]

241. In England, the issues have also been considered. In Hodgson v Imperial Tobacco Ltd,[174] Lord Woolf took the view that, even when there is a private hearing, the public continues to have the right to know and observe what happens in that private hearing and arrangements need to be made to accommodate that right.

The principles as they apply to a statutory tribunal

242. These are but a handful of the authorities that consider the issue but they set out the fundamental principles. They are that:

(1) a private hearing, whether required by legislation or ordered by a court or tribunal with appropriate power:

(b) does not of itself impose any restrictions on the publication of the evidence given at, or material or information relating to that private hearing;

(a) specify the consequences of a private hearing;[175] and/or

(b) confer power, either expressly or implicitly, on the court or tribunal to make orders specifying the consequences;[176]

(3) a tribunal such as the Administrative Appeals Tribunal, whose power is conferred by statute:

(a) has no power to impose any restrictions on publication unless:

(i) the power to do so has been expressly, or by necessary implication, conferred upon it; and

(ii) it has exercised that power within the bounds upon which it has been conferred;

(4) the matters relevant to Parliament's deciding that proceedings should be conducted in private in a court or tribunal's ordering that they be conducted in private may not equate precisely or at all with the matters that are relevant to whether an order should be made restricting publication of those proceedings.[177]

243. In drafting s 39T(4) of the IRD Act, the Parliamentary draftsman appears to have been guided by these principles. After stating that the hearing must be in private, the provision goes on:

"... the Administrative Appeals Tribunal may, by order:

(a) give directions as to the persons who may be present; and

(b) give directions of a kind referred to in paragraph 35(2)(b) or (c) of the Administrative Appeals Tribunal Act 1975."

244. Section 39T(4)(a) is self-explanatory but ss 35(2)(b) and (c) of the AAT Act to which s 39T(4)(b) refers, are not. They provide:

"Where the Tribunal is satisfied that it is desirable to do so by reason of the confidential nature of any evidence or matter or for any other reason, the Tribunal may, by order:

(b) give directions prohibiting or restricting the publication of evidence given before the Tribunal, whether in public or in private, or of matters contained in documents lodged with the Tribunal or received in evidence by the Tribunal; and

(c) give directions prohibiting or restricting the disclosure to some or all of the parties to a proceeding of evidence given before the Tribunal, or of the contents of a document lodged with the Tribunal or received in evidence by the Tribunal, in relation to the proceeding."

245. No reference is made to directions that the Tribunal might give under ss 35(2)(a) and (aa) for they refer to whether the hearing, or part of it, shall be in private and those who may be present as well as to restricting or prohibiting the publication of the names and addresses of witnesses. The matters dealt with in s 35(2)(a) are already the subject of s 39T(4). Arguably, the power to restrict or prohibit the publication of the the names and addresses of the witnesses is encompassed within the power to prohibit or restrict the publication of evidence for those details will themselves be a matter of evidence whether given in written or verbal form.

246. When the Tribunal makes directions of the sort provided for in ss 35(2)(b) or (c), it must do so having regard to the matters set out in s 35(3) of the AAT Act.[178] It does not follow that those matters are relevant in considering whether to make an order when considering an application for review of a decision under the IRD Act. It does not follow because in proceedings relating to the review of a decision under the IRD Act, ss 35(2)(b) and (c) of the AAT Act do not confer power on the Tribunal to make the orders. The power is conferred by s 39T(4)(b) of the IRD Act. That provision is very specific in providing that the Tribunal may give directions "of a kind referred to in paragraph 35(2)(b) or (c) ..." (emphasis added) of the AAT Act. That is a power to make directions "of the same sort"[179] as those made under ss 35(2)(b) or (c) but not a power to make them under those provisions.

247. That is not to say that some of the matters of the kind referred to in s 35(3) of the AAT Act are not relevant even though the section itself is not relevant. The matters are not relevant in so far as they refer to its being desirable that hearings of proceedings in the Tribunal shall be held in public. Parliament has already decided that hearings of proceedings relating to the review of decisions made under the IRD Act are to be held in private.

248. It is clear from Scott v Scott that the remaining matters referred to in s 35(3) are among those that continue to be relevant. They are that:

"The Tribunal shall take as the basis of its consideration the principle that it is desirable ... that evidence given before the Tribunal and the contents of the documents lodged with the Tribunal or received in evidence by the Tribunal should be made available to the public and to all parties, but shall pay due regard to any reasons given to the Tribunal why the hearing should be held in private or why publication or disclosure of the evidence or the matter contained in the document should be prohibited or restricted."

249. These matters would be moulded by considerations arising from the nature of the evidence and material in a particular case and a more detailed consideration of the IRD Act. We do not need to engage in that as RACV Sales has not asked that an order be made protecting its identity. It should be identified by its name.

I certify that the preceding two hundred and forty nine paragraphs are a true copy of the reasons for the decision herein of

Deputy President S A Forgie and Senior Member E Fice.

Signed: ....................................................................

[1] Income Tax Assessment Act 1936 (ITAA36), s 73B(14)

[2] RACV Sales was identified in the Tribunal's records by the letters "KSLZ". No order, however, has been applied for, or made, under s 35 of the Administrative Appeals Tribunal 1975 (AAT Act) restricting publication of RACV Sales' identity. Section 39T(4) of the IRD Act requires that the Tribunal's hearing of a proceeding relating to a decision under the IRD Act must take place in private. Without an order prohibiting disclosure, a private hearing does not, of itself, prohibit or restrict disclosure of what is said and the evidence that is given at that private hearing. Therefore, we have identified the applicant. The parties had no objection to our doing so and, indeed, Innovation Australia asked that the applicant's name be disclosed. We explain our reasons more fully in Attachment C.

[3] All but the first two years are references to financial years. The first two relate to slightly different periods but the differences are of no consequence in this case. Although there is no document recording registration, the statement prepared by the Board under s 37(1)(a) of the AAT Act states that "The applications for registration were accepted by the Respondent for these years.": T documents at 3.8, [18]. The registered activities are described as the RSM-1 Project but, in 2005, the project became the RACV Sales 05-01 Project. Nothing turns on this change of name.

[4] References in the IRD Act to the "Board" are references to Innovation Australia: IRD Act, s 4(1).

[5] There is no material in the documents we have been given that shows that the Board has registered RACV Sales in relation to the research and development activities in relation to which it has applied for registration under s 39J. Apart from the issue relating to the characterisation of RACV Sales' activities, the Board was not satisfied that RACV Sales has given it material showing that RACV Sales had "... while the activities were carried on, maintained records that substantiate the carrying on of the activities": IRD Act, s 39JD(ba). The Board noted at its meeting on 27 July 2009 that it had also asked RACV Sales for information on the RSM1 project and, if necessary, a technical opinion following its meeting held in March 2007:T documents, T3 at 6. Earlier communications regarding that material took place in November 2005: T documents, T37 and T38 at 769-775.

[6] Section 73B of ITAA36 was repealed with effect from 8 September 2011: Tax Laws Amendment (Research and Development) Act 2011; Act No 93 of 2011, s 2, item 6 and s 3, Schedule 3, Part 6, item 44. The application of s 73B and the relevant provisions of the IRD Act such as s 39L is continued in relation to acts done or committed and the state of affairs existing before 8 September 2011; Act No 93 of 2011, s 3, Schedule 4.

[7] A more detailed description appears at [7] below.

[8] T documents at 6 and ST documents at 1493. We consider these decisions in more detail at [39]-[44] below

[10] Mr Case is the Manager, Vehicle Engineering, with RACV Sales

[11] Participation Agreement dated 22 December 1992, Attachment 2, item A: Exhibit C at MC 2

[13] T documents at 1288 and Exhibit C at [2.5]

[14] Consortium Agreement, Exhibit C at MC 4, 2

[15] Supplementary Documents lodged under s 37 of the Administrative Appeals Tribunal Act 1975 (ST documents) at 1493. The Board added by way of comment:

"Based on the information available the activities undertaken in the project, 'RSM1 - Understanding the Results of Crash Testing and the Impact of Automotive Structural Design and Safety', carried out in the 1998-99 to 2005-06 years of income were not systematic, investigative and experimental." (ST Documents at 1493)

[16] Respondent's Statement of Facts, Issues and Contentions at [6.2]

[17] During the course of the hearing and subsequently, there has been some disquiet between the parties regarding some matters that have been raised and the manner and time at which they have been raised. We have addressed these at Attachment B.

[19] That person would have to be registered in respect of those activities in order to claim a deduction under s 73B of ITAA36.

[24] Industry Research Board v Unisys Information Services Australia Ltd (1997) 77 FCR 552; 37 ATR 62; Industry Research Board v Coal & Allied Operations Pty Ltd [2000] FCA 979; (2000) 101 FCR 405; 44 ATR 541; Re Charles IFE Pty Ltd and Industry Research and Development Board [1995] AATA 224; (1995) 39 ALD 635; 95 ATC 2149; [1995] AATA 224; 32 ATR 1226; Re Fermenter and Distiller and Industry Research and Development Board [2000] AATA 888; (2000) 45 ATR 1122 and Re The Applicant and Industry Research and Development Board [1999] AATA 468; (1999) 55 ALD 784; 42 ATR 1116

[27] Taxation Laws Amendment (Research and Development) Act 2001, Act No. 170 of 2001, s 3, Schedule 1, item 1. It was amended in 2007 but that was at a time outside the years we must consider.

[28] With effect from 25 September 2007, s 73B(1AAA) was amended by adding after the words "deduction to" the words "encourage research and development activities in Australia and": Tax Laws Amendment (2007 Measures No. 5) Act 2007, Act No. 164 of 2007, s 3, Schedule 11, item1.

[29] Section 73B(1AAA)(c) was enacted in the form of "innovation or high technical risk" (emphasis added). It was not amended in 2007 when the words "encourage research and development activities in Australia and" (emphasis added) were inserted in the opening words of s 73B(1AAA); see preceding FN. We note that the CCH publication "Australian Income Tax Legislation" and its loose leaf service incorrectly records s 73B(1AAA)(c) as reading "innovation and high levels of technical risk" (emphasis added).

[33] Section 39KA was inserted by the Taxation Laws Amendment (Research and Development) Act 2001, Act No. 170 of 2001, s 3, Schedule 1, item 7. With effect from 1 July 2002, s 73B(2BA) was inserted in ITAA36: Act No. 170 of 2001, s 3, Schedule 1, items 5 and 6.

[34] "For the purposes of the definition of 'research and development activities' in subsection (1), activities carried on by or on behalf of an eligible company by way of development of computer software shall not be taken to be systematic, investigative and experimental activities unless the computer software is developed for the purpose, or for purposes that include the purpose, of sale, rent, licence, hire or lease to two or more non-associates of the company ...".

[36] The exclusions identified in s 73B(2C) are also relevant in identifying a "program of systematic, investigative and experimental activities" for the purposes of s 73B(2B)(b); see [163] and [165] below

[38] Taxation Laws Amendment (Research and Development) Act 2001, Act No. 170 of 2001, s 3, Schedule 1, item 2

[39] During the years in issue in this case, advance registration was provided for eligible companies in relation to "proposed research and development activities" (emphasis added). Section 39HG set out the criteria required for advance registration. Subject to s 39HI, if an eligible company complied with those criteria and provided the Board with such information in relation to the proposed research and development activities as it reasonably required, the Board was required to grant advance registration to the company in relation to those activities and in respect of the years of income to which the application related: IRD Act, s 39HH(1). Section 39HI(1) provided that:

"It is a ground on which the Board is entitled to refuse advance registration of an eligible company, in relation to particular research and development activities in respect of a year or years of income, that the activities are not research and development activities."

Regulations might specify other grounds on which the Board was entitled to refuse advance registration provided those grounds related to a matter that was within the functions of the Board to determine: IRD Act, ss 39HI(2) and (3).

[44] Section 39N(3) provides that the written notice must also state that, if the company fails to comply with the notice, the Board may give a certificate under s 39N(2). It must also set out the terms of ss 73B(33A), 73BF(4) and 73BM(4) of ITAA36.

[46] Section 139L(3) is concerned with activities carried on by or on behalf of a foreign company grouped under s 73L of ITAA36 with an eligible company. It is not relevant.

[47] Act No 59 of 1988, ss 5(d) and (e) and 16

[48] Taxation Laws Amendment (No 4) 1989, Act No 167 of 1989, s 34

[49] ITAA36, s 124ZG(2A) as amended by Act No. 167 of 1989, s 34

[51] Industry Research and Development Act 1989, Act No. 82 of 1996, s 3, Schedule 1, item 12

[52] Chambers 21st Century Dictionary, 1999, reprinted 2004, Chambers (Chambers)

[54] [2000] AATA 888; (2000) 45 ATR 1122

[55] [2000] AATA 888; (2000) 45 ATR 1122 at [15]; 1126

[57] The definition was amended by the Taxation Laws Amendment Act (No 3) Act 1996, Act No. 78 of 1996, s 3 and Schedule 4, item 53.

[58] [1995] AATA 224; (1995) 39 ALD 635; 95 ATC 2149; [1995] AATA 224; 32 ATR 1226; Deputy President McDonald and DL Elsum and RC Gillham, Members

[59] [1995] AATA 224; (1995) 39 ALD 635; 95 ATC 2149; [1995] AATA 224; 32 ATR 1226 at [27]; 643; 2,155-2,156; 1233

[60] (1998) 85 FCR 382; 52 ALD 246; 28 AAR 36

[61] (1998) 85 FCR 382; 52 ALD 246; 28 AAR 36 at 385; 249; 39

[62] Justice Drummond acknowledged that "... the Tribunal referred to judicial warnings against applying to expressions in the Social Security Act the meanings given to similar expressions in other statutes, particularly the Income Tax Assessment Act 1936 (Cth) (ITAA). However, there is, in my opinion, good reasons to read the expression 'carries on a business' in s 1075 of the Social Security Act as having exactly the same meaning as the expression 'carrying on a business' in s 51(1) of the ITAA. ...". (1998) 85 FCR 382; 52 ALD 246; 28 AAR 36 at 386; 249; 39-40. He went on to trace the history of the relevant provisions to show why it should be so.

[67] The word "systematic", for example, has been interpreted in the context of considering whether employment is "regular and systematic". Employment can be regular and systematic even if the breaks in that employment and the hours worked do not reflect any mathematical consistency. What is relevant is the whole of the employment relationship including matters such as the nature of the work, constants in terms of nominated hours and allocated duties, mutual accommodation of periods of leave or periods when work is not available, notice of such periods to the other, at the instigation of the employer and the period over which such arrangements have existed between an employer and employee. See, for example, Heggie v Minda Incorporated [2006] SAIRComm 9 at [42]- [57]; Commissioner D Steel

[69] Section 30 of the Personal Injuries Proceedings Act 2002 (Qld) (PIPA) provides that a party is not obliged to disclose documentary material subject to legal professional privilege but does not exclude from disclosure "investigative reports, medical reports and reports relevant to the claimant's rehabilitation ... with the omission of passages consisting only of statements of opinion." The purpose of PIPA was "to put the parties in a position where they have enough information to assess liability and quantum in relation to a claim". Fraser JA said in Queensland v Allen [2011] QCA 311; Fraser and White JJA and Fryberg J at [26]-[27]:

"26. ... I accept that the process in which PCH [Prince Charles Hospital) engaged upon the advice of its solicitor was a systematic enquiry of the doctors who might be able to provide information about the medical procedure and related matters. That overall process might be regarded as an 'investigation' within the ordinary meaning of that word, but the question is whether a particular document is itself an 'investigative report'. None of the documents in issue has that quality. None is a report of the result of the overall process recommended by the solicitor. Rather, each doctor separately prepared (or separately supplied information to a solicitor who prepared) a record of that doctor's recollections of and, in some cases, opinions about, the procedure and related matters. No doctor was asked to conduct any investigation or research for the purpose of expressing an opinion or otherwise. The resulting record of information is no more 'investigative' in character than any witness statement or solicitor's file note of information within the witness' own knowledge. In a particular case, such a record might be attached to a broader report of an investigation by, for example, a loss assessor, and thereby form part of that 'investigative report',... but there was no such report here. Each document was simply retained by PCH in its files for subsequent use in the anticipated litigation.

27. In my opinion, a statement by a witness to an incident alleged to have caused personal injury to a claimant, or a solicitor's file note, which records that person's recollection of the circumstances of the incident and the person's opinion about the incident for use in anticipated litigation, is not, in ordinary parlance, an 'investigative report'. Acceptance of the respondent's submission to the contrary would result in the widespread abrogation by s 30(2) of privilege in witness statements taken by solicitors for use in existing or anticipated litigation or for the purpose of giving confidential legal advice, since every witness statement should be the product of focused and systematic enquiries by solicitors. There is no indication in s 30 that such a result was intended." (citations omitted)

[70] Section 9(a) of the Patents Act 1900, for example, provided that "For the purposes of this Act, the following acts are not to be taken to be secret use of an invention in the patent area: (a) any use of the invention by or on behalf of, or with the authority of, the patentee or nominated person, or his or her predecessor in title to the invention, for the purpose of reasonable trial or experiment only; ..." In Azuko Pty Ltd v Old Digger Pty Ltd [2001] FCA 1079; (2001) 52 IPR 75; [2001] AIPC 91-741

(Beaumont, Heerey and Gyles JJ) Heerey J concluded that:

"147 The evidence permitted of only one conclusion. Production of fifteen or twenty hammers in a condition ready for commercial sale, an order having been received for five or six, cannot be considered a matter of trial and experiment, let alone only trial and experiment. ...

148 There is a total lack of evidence from Mr Giehl as to what trials or experiments were carried out on the Adelaide hammers. The reasonable inference is that there were none. Even if he were making a limited number to see if production was commercially viable, I do not think that would be the kind of trial and experiment of which s 9(a) speaks. The provision is limited to trial or experiment to see how the product of an invention performs and whether any improvements are needed, as distinct from commercial or marketing assessments. But in any event Mr Giehl did not suggest his manufacture was for the limited purpose found by his Honour. The inference to be drawn is simply that Mr Giehl manufactured hammers because he proposed to sell them."

[71] [1944] HCA 4; (1944) 68 CLR 261; Latham CJ, Rich, Starke, McTiernan and Williams JJ

[72] [1944] HCA 4; (1944) 68 CLR 261 at 274-275 (citations omitted)

[74] (1997) 77 FCR 552; 37 ATR 62; Carr, RD Nicholson and Finn JJ

[75] [2000] FCA 979; (2000) 101 FCR 405; 44 ATR 541

[76] [2000] FCA 979; (2000) 101 FCR 405; 44 ATR 541 at [53]- [54]; 421; 556

[78] [1993] AATA 254; (1993) 30 ALD 200; 18 AAR 177; 26 ATR 1262; Deputy President Thompson and Members DL Elsum and WG McLean

[79] [1993] AATA 254; (1993) 30 ALD 200; 18 AAR 177; 26 ATR 1262 at [40]; 210; 188; 1272

[80] [1993] AATA 254; (1993) 30 ALD 200; 18 AAR 177; 26 ATR 1262 at [45]; 2121; 189-190; 1274

[81] The Second Reading Speech referred to the definition of "research and development activities" having been "... formulated after extensive consultation with industry, the scientific community and the Australian accounting profession. It is based on the definition of R&D used by the OECD and the Australian Bureau of Statistics. It closely resembles the definition used by the Canadian Government which operates a taxation concession scheme with similar features to this proposed concession." (emphasis added) Hansard, Senate, 2 June 1986 at 3152

[82] Hansard, Senate, 2 June 1986 at 3152 The Explanatory Memorandum state that "A basic criterion associated with determination of 'innovation' is the presence of an appreciable element of novelty. This may be assessed according to several criteria, including whether - • the work is seeking previously undiscovered phenomena, structures or relationships; • the work is attempting to apply knowledge or techniques in a new way; • the outcomes are likely to benefit more than one organisation; or • the results are expected to be patentable."

[83] [1997] AATA 67; Deputy President McMahon and Mr Way, Member

[84] [1997] AATA 67 at [34]- [35]

[85] Unisys (1997) 77 FCR 552; 37 ATR 62 at 558-559; 68

[86] Taxation Laws Amendment Act (No 3) 1996, Act No 78 of 1996, s 3 and Schedule 4, Part 1, Division 7, item 56

[87] (1997) 77 FCR 552; 37 ATR 62 at 558; 67

[88] "The Latin sentence de minimis non curat lex means the law is not concerned with trifles.": Ross on Crime 5th edition, Lawbook Co, 2011 at 483

[89] The Explanatory Memorandum was revised but the revisions did not relate to the provisions with which we are concerned.

[90] Re Bolton; Ex parte Beane [1987] HCA 12; (1987) 162 CLR 514; 70 ALR 225; 61 ALJR 190 at [4]; 518; 227-228; 191 per Mason CJ, Wilson and Dawson JJ

[91] [1988] FCA 72; (1987) 11 NSWLR 126; [1988] ASC 57; Kirby P, Mahoney and Priestley JJA at 128-129; 57,782-57,783

[92] [1988] FCA 72; (1987) 11 NSWLR 126; [1988] ASC 57 at 128; 57,782

[94] [1997] AATA 67 at [16]

[95] (1997) 77 FCR 552; 37 ATR 62 at 559; 68-69

[96] [2000] FCA 979; 101 FCR 405; 44 ATR 541 at [80]- [81]; 426; 560

[97] [1999] AATA 468; (1999) 55 ALD 784; 42 ATR 1116

[98] [2010] AATA 879; Deputy President McDonald and Mr C Ermert, Member

[99] [2010] AATA 879 at [37]

[100] Those appearing for HZXD had no record of the work that had been done as the hard drive on a computer had been damaged and they could not afford to spend $5,000 to retrieve it. No back-up records had been maintained. [2010] AATA 879 at [18]- [19]

[101] See generally Re Australian Federation of Construction Contractors; Ex parte Billing [1986] HCA 74; (1986) 68 ALR 416; (1986) 61 ALJR 37 at [4]; 420; 39 per Gibbs CJ Mason, Wilson, Brennan, Deane and Dawson JJ

[104] DesignsAct 2003, the Olympic Insignia Protection Act 1987, the Patents Act 1990, the Plant Breeder's Rights Act 1994 and the Trade Marks Act 1995

[105] An "invention means any manner of new manufacture the subject of letters patent and grant of privilege within section 6 of the Statute of Monopolies, and includes an alleged invention." Patents Act, Schedule 1 and see s 3.

[106] Patents Act, s 18(1)(b)(i). That is explained further in s 7(1): "For the purposes of this Act, an invention is taken to be novel when compared with the prior art base unless it is not novel in light of any one of the following kinds of information, each of which must be considered separately: (a) prior art information (other than that mentioned in paragraph (c)) made publicly available in a single document or through a single act; (b) prior art information (other than that mentioned in paragraph (c)) made publicly available in 2 or more related documents, or through doing 2 or more related acts, if the relationship between the documents or acts is such that a person skilled in the relevant art would treat them as a single source of that information; (c) prior art information contained in a single specification of the kind mentioned in subparagraph (b)(ii) of the definition of prior art base is Schedule 1."

[107] Patents Act, ss 18(1)(c) and (d) and 18(1A)(c) and (d)

[112] We find ourselves in disagreement with the explanation of s 73B(2A)(a) given in [9.56] of the Explanatory Memorandum and reproduced at [188] above. It ends with the statement that "This means that a fairly large constituent part of the activity must involve novelty." That would seem to suggest that the adjective "appreciable" is the descriptor of the word "activities". For the reasons we have given, it is not. Having attributed it in this way, the statement then suggests that activities can be further divided and an assessment made of each part so that a decision can be made as to whether "a fairly large constituent part of the activity" involves novelty. Only if it does can it be said to have an appreciable element of novelty. On our understanding of the provision, the wording of the provision requires us to look at the activities as a whole and decide whether they have a component or feature of being new. That follows from the section itself and was underlined by Lindgren J in Coal & Allied when he said that "... According to either meaning, an activity, although it must be 'particular', is not disqualified by reason only of the fact that in some respects it does not involve innovation or technical risk. ..." (see [168]-[169] above. It may be that, in a particular instance, a finding of innovation will only be made if "a fairly large constituent part of the activity" involves novelty. It may equally be that, when each constituent part is assessed, very few involve novelty on their own but, when inter-linked and viewed as a whole, come within the description of "activities ... [that] involve an appreciable element of novelty." All will depend on the evidence in a particular case.

[113] [1997] AATA 67

[114] [1999] AATA 468; (1999) 55 ALD 784; 42 ATR 1116 at [26(i)]

[115] [2003] FCAFC 143; (2003) 131 FCR 28; 37 AAR 309; Gray ACJ and North; Gyles JJ

[116] [2003] FCAFC 143; (2003) 131 FCR 28; 37 AAR 309 at [61]- [62]; 45; 325

[117] In this regard, the Court's duty differs markedly from that of the Tribunal. The Tribunal's duty is to reach the correct decision on both facts and law and the preferable decision if it is open to make more than one correct decision; see generally Drake v Minister for Immigration and Ethnic Affairs (1979) 2 ALD 60; 24 ALR 577; 46 FLR 409 at 68; 589; per Bowen CJ and Deane J. In the same case, Smithers J said: "The duty of the Tribunal is to satisfy itself whether a decision in respect of which an application for review is duly instituted is a decision which in its view, was objectively, the right one to be made." (1979) 2 ALD 60; 24 ALR 577; 46 FLR 409 at 77; 599; 429-430. Unlike the Court which must confine itself to the grounds of appeal and so the legal issues raised by the parties, the Tribunal generally has a duty to consider issues raised by the legislation and the material even when they have not been raised by the parties. If, for example, there was a clear statutory precondition upon which the Tribunal had to be satisfied and enough material and evidence before it to raise the issue independently of the parties' submissions, the Tribunal would make an error of law were it not to do so: see, for example, Kuswardana v Minister for Immigration and Ethnic Affairs [1981] FCA 66; (1981) 54 FLR 334; 35 ALR 186 at 341-342; 194 per Bowen CJ and at 348; 199 per Fox J. ; McKeown v Repatriation Commission (1995) 39 ALD 30; 22 AAR 229 at 33; 232 per Jenkinson J; Australian Trade Commission v F & F Asia Pty Ltd (1996) 69 FCR 252; 42 ALD 197 at 266; 209. per Carr J; and Transport Accident Commission v Bausch [1998] 4 VR 249; 13 VAR 61 at 76; 263 per Tadgell JA; Batt and Buchanan JJA agreeing.

[118] (1997) 77 FCR 552; 37 ATR 62 at 561; 70

[119] (1997) 77 FCR 552; 37 ATR 62 at 561; 70

[120] (1997) 77 FCR 552; 37 ATR 62 at 561-562; 70-71

[121] [1997] AATA 67 at [37]

[122] Set out at [199] above but repeated here: "_The statement that it is the process not the product that must meet the criteria is not correct. There is nothing in the legislation requiring that the definition be read down to refer only to processes rather than products (_Unisys at first instance not dissented from by the Full Court)."

[123] [1997] AATA 67 at [37]

[124] [1959] HCA 8; (1985) 101 CLR 298; 32 ALJR 395; Kitto, Menzies and Windeyer; Dixon CJ and Taylor J dissenting

[125] [1959] HCA 8; (1985) 101 CLR 298; 32 ALJR 395 at 305; 397. Although in dissent, the following passage from the judgment of Dixon CJ explains the principle. A truck driver who was killed when, travelling uphill, his International truck came into collision with an unladen diesel truck travelling downhill in the opposite direction. His widow sued for the owner and driver of the unladen truck for damages alleging that the driver had been negligent. His Honour said:

"... The accident is simply left unexplained. No doubt the conclusion is reasonably open that at the moment of actual impact the right side of the forward part of the International truck and the front of the diesel truck were in violent contact. The inference is also open that the diesel truck was swung round to face the other way. Strange as such an effect of the forces may seem, perhaps it might also be reasonably concluded that the left-hand door flew open and was torn off as the truck went round, notwithstanding the absence of evidence of marks upon any of the posts at the edge of the road. But the all important question of the cause of the vehicles hitting one another is left unresolved by the circumstantial evidence. It is possible of course to say that if you have an empty diesel truck coming down a winding road on the outside at thirty-five miles per hour and an International truck going up the road on the inside at twenty-five miles per hour, the former is more likely than the latter to be over the centre line of the road on its wrong side. But that is only to say that of two guesses one is more probable than another. It may be remarked that these are not the only two guesses open as to the cause of the accident. But in any case we are not concerned with a choice among rival conjectures. In an action of negligence for death or personal injuries the plaintiff must fail unless he offers evidence supporting some positive inference implying negligence and it must be an inference which arises as an affirmative conclusion from the circumstances proved in evidence and one which they establish to the reasonable satisfaction of a judicial mind. It is true that 'you need only circumstances raising a more probable inference in favour of what is alleged'. But 'they must do more than give rise to conflicting inferences of equal degree of probability so that the choice between them is mere matter of conjecture'. These phrases are taken from an unreported judgment of this Court in Bradshaw v. McEwans Pty. Ltd. ... which is referred to in Holloway v. McFeeters ..., by Williams, Webb and Taylor JJ. The passage continues: 'All that is necessary is that according to the course of common experience the more probable inference from the circumstances that sufficiently appear by evidence or admission, left unexplained, should be that the injury arose from the defendant's negligence. By more probable is meant no more than that upon a balance of probabilities such an inference might reasonably be considered to have some greater degree of likelihood.' ... But the law which this passage attempts to explain does not authorise a court to choose between guesses, where the possibilities are not unlimited, on the ground that one guess seems more likely than another or the others. The facts proved must form a reasonable basis for a definite conclusion affirmatively drawn of the truth of which the tribunal of fact may reasonably be satisfied." [1959] HCA 8; (1985) 101 CLR 298; 32 ALJR 395 at [2]; 305; 396-397 (citations omitted)

[126] [1995] AATA 133; (1995) 95 ATC 2042; 30 ATR 1364; Deputy President Forrest, D Elsum and R Gillham, Members

[127] [1995] AATA 133; (1995) 95 ATC 2042; 30 ATR 1364 at [34]; 2047; 1369

[128] Whitten v Falkiner [1915] HCA 38; (1915) 20 CLR 118 at 127 per Isaacs J and see also HR Products Pty Ltd v Collector of Customs [1990] FCA 152; (1990) 20 ALD 340 at 342 per Lee J and Pepsi Seven-Up Bottlers v Commissioner of Taxation (1995) 62 FCR 289; 132 ALR 632 at 299; 641-642 per Hill J

[129] (1995) 62 FCR 289; 132 ALR 632 at 298-299; 641

[130] [1999] AATA 468; (1999) 55 ALD 784; 42 ATR 1116

[131] [1999] AATA 468; (1999) 55 ALD 784; 42 ATR 1116 at [26]; 791-792; 1123

[132] "Mr Forsyth submitted that the refinery trial whereby ALC was co-mingled with GC was innovative because it had not been done before by anyone or, if it had, it had been held in confidence. He argued there was technical risk because the applicant did not know what products would eventuate since complex interactions take place; for example, would an admixture of ALC in the crude feed materially lower the pour point of products, in particular HFO, recognising that, as he said, the pour point of GC is 42 degrees C and that of ALC is -15 degrees C? So far as sulphur was concerned, the question was where would the additional sulphur in the co-mingled crude end up? Would it be spread over the products and how would it affect the air emissions? He said the applicant went to a great deal of trouble, involving getting the ALC, organising tankage, communicating with a major customer and the Environment Protection Authority, carrying out more frequent sampling of refinery products, making constant adjustments to operating parameters and rerouting streams during the refinery trial. The applicant wanted to know what would happen when it mixed ALC with GC." 133 [1995] AATA 133; (1995) 95 ATC 2042; 30 ATR 1364 at [26]; 1046; 1368

[134] [1995] AATA 133; (1995) 95 ATC 2042; 30 ATR 1364 at [35]-[36]; 2,047; 1369

[135] (1997) 77 FCR 552; 37 ATR 62 at 559; 68

[136] (1997) 77 FCR 552; 37 ATR 62 at 560-561; 69-70

[137] (1997) 77 FCR 552; 37 ATR 62 at 560-561; 70

[138] [2000] FCA 979; (2000) 101 FCR 405; 44 ATR 541 at [98]; 429; 563

[143] Macquarie Dictionary, revised 3rd edition, 2001, The Macquarie Library Pty Ltd

[145] The word "uncertainty" is a noun and is derived from the word "uncertain", which means, in so far as it is relevant in this context: "... 1 not sure, certain or confident. 2 not definitely known or decided. ..." (Chambers).

[146] See [156]-[165] above and summary at [24(3)]

[157] Second Reading Speech, Income Tax Assessment Amendment (Research and Development) Bill 1986, Hansard, Senate, 2 June 1986 at 3151

[158] [2000] FCA 979; (2000) 101 FCR 405; 44 ATR 541 at [103]; 430; 564 and see also [108]; 431; 565

[159] [2000] FCA 979; (2000) 101 FCR 405; 44 ATR 541 at [106]; 431; 564

[160] Blacktown Workers' Club Ltd v O'Shannessy [2011] NSWCA 265; (2011) 183 LGERA 184; Basten JA, Handley and Sackville AJJA at [52]; 197 per Basten JA

[163] [1989] FCA 530; (1989) 24 FCR 217; 91 ALR 321; Lockhart, Beaumont and Einfeld JJ

[164] [1989] FCA 530; (1989) 24 FCR 217; 91 ALR 321 at 221; 325

[165] [1913] AC 417; [1911-1913] All ER 1

[166] [1913] AC 417; [1911-1913] All ER 1 at 437-438; 9

[167] [1913] AC 417; [1911-1913] All ER 1 at 442 and 443; 12

[168] [1913] AC 417; [1911-1913] All ER 1 at 447; 14

[169] [1913] AC 417; [1911-1913] All ER 1 at 453; 17

[170] [1913] AC 417; [1911-1913] All ER 1 at 476; 29

[171] (1988) 14 NSWLR 342; Hope, Mahoney and McHugh JJA

[172] (1988) 14 NSWLR 342 at 345-346

[173] Section 36(1) provided that the Commission might hold hearings for the purpose of performing its functions. Section 36(2) provided that it might direct that the hearing take place in public or in private. Where a private hearing was held, s 36(5) permitted the Commission to direct those who might be present at the hearing. Where a public hearing was held, s 36(6) permitted the Commission to direct that part might take place in private, direct those who might be present and prevent or restrict the publication of the evidence.

[174] [1998] EWCA Civ 224; [1998] 2 All ER 673 at 686

[175] The IRD Act does not provide for any consequences but an enactment such as the Taxation Administration Act 1953 (TA Act) does. Section 14ZZE provides that:

"Despite section 35 of the AAT Act, the hearing of a proceeding before the Tribunal, other than the Small Taxation Claims Tribunal, for:

(a) a review of a reviewable objection decision; or

(b) a review of an extension of time refusal decision; or

is to be in private if the party who made the application requests that it be in private."

The consequence is specified in s 14ZZJ. It provides that s 43 of the AAT Act is to be read as if it included ss 43(2C), (2D) and (2E). Their effect is that the fact that a hearing of a proceeding is conducted in private does not prevent the Tribunal from publishing its reasons for decision but it must ensure, as far as practicable, that its reasons are framed so as not to be likely to enable the identification of the person who applied for review.

[176] The IRD Act does so expressly but an Act such as the TA Act does so implicitly for there is no specific reference to the Tribunal's powers under s 35 of the AAT Act. Justice Emmett took this view in

in Brown v Commissioner of Taxation [2001] FCA 276; (2001) 47 ATR 143 at 145-146 when he said that: "... it would be a most unusual case where the tribunal, if asked, did not give directions that are contemplated by s 35(2) in a proceeding to which s 14ZZE applies. The tribunal is empowered to give directions for any reason, where it is satisfied that it is desirable to do so. Where a party exercised the right, under s 14ZZE, to have a hearing in private, that would be a very cogent reason for the tribunal to make an order under s 35(2)(b)."

[177] It is clear from the passage from Lord Loreburn's speech in Scott v Scott that there are many reasons for a private hearing but not all of them would justify a non-publication order:

" It has been held that when the subject-matter of the action would be destroyed by a hearing in open Court, as in a case of some secret process of manufacture, the doors may be closed. I think this may be justified upon wider ground. Farwell LJ aptly cites Lord Eldon as saying, in a case of quite a different kind, that he dispensed with the presence of some of the parties 'in order to do all that can be done for the purposes of justice rather than hold that no justice shall subsist among the persons who may have entered into these contracts.' An aggrieved person, entitled to protection against one man who had stolen his secret, would not ask for it on the terms that the secret was to be communicated to all the world. There would be in effect a denial of justice.

Again, the Court may be closed or cleared if such a precaution is necessary for the administration of justice. Tumult or disorder, or the just apprehension of it, would certainly justify the exclusion of all from whom such interruption is expected, and, if discrimination is impracticable, the exclusion of the public in general. Or witnesses may be ordered to withdraw, lest they trim their evidence by hearing the evidence of others. Or, to use the language of Fletcher Moulton LJ, in very exceptional cases ... where a judge finds a portion of the trial is rendered impracticable by the presence of the public, he may exclude them so far as to enable the trial to proceed. It would be impossible to enumerate or anticipate all possible contingencies, but in all cases where the public has been excluded with admitted propriety the underlying principle, as it seems to me, is that the administration of justice would be rendered impracticable by their presence, whether because the case could not be effectively tried, or the parties entitled to justice would be reasonably deterred from seeking it in the hands of the Court." [1913] AC 417; [1911-1913] All ER 1 at 445-446; 3-14

Lord Woolf has discussed the issues more recently in Hodgson v Imperial Tobacco Ltd [1998] EWCA Civ 224; [1998] 2 All ER 673 at 686. He said that a hearing in private may contribute:

"... to the administration of justice. They allow issues to be determined informally and expeditiously. They allow less strict rules as to representation to apply. They allow matters to be discussed which the parties might not wish to discuss in open court. They encourage openness. They are less intimidating to litigants which is particularly important in the case of the small claims jurisdiction ..."

[178] "... the Tribunal shall take as the basis of its consideration the principle that it is desirable that hearings of proceedings before the Tribunal shall be held in public and that evidence given before the Tribunal and the contents of documents lodged with the Tribunal or received in evidence by the Tribunal should be made available to the public and to all the parties, but shall pay due regard to any reasons given to the Tribunal why the hearing should be held in private or why publication or disclosure of the evidence or the matter contained in the document should be prohibited or restricted."