Ultimate Vision Inventions Pty Ltd v Innovation and Science Australia

The ITAA 1997 7 As I have mentioned, the relevant provisions of the ITAA 1997 are contained in Division 355 (Research and Development). Section 355-1 explains that an R&D entity may be entitled to a tax offset for R&D activities. To be entitled to the tax offset, the R&D entity needs one or more notional deductions under the Division. Section 355-1 also explains that there are two kinds of notional deductions: one is for expenditure on R&D activities; the other is for the decline in value of tangible depreciating assets used for R&D activities. A note under s 355-1 states that all of these notional deductions require the R&D entity to be registered for the R&D activities under Pt III of the IR&D Act. 8 The term "R&D entity" is defined in s 355-35. The definition includes a body corporate incorporated under Australian law. 9 The object of Division 355 is set out in s 355-5 - 355-5 Object (1) The object of this Division is to encourage industry to conduct research and development activities that might otherwise not be conducted because of an uncertain return from the activities, in cases where the knowledge gained is likely to benefit the wider Australian economy. (2) This object is to be achieved by providing a tax incentive for industry to conduct, in a scientific way, experimental activities for the purpose of generating new knowledge or information in either a general or applied form (including new knowledge in the form of new or improved materials, products, devices, processes or services). 10 The expression "R&D activities" is defined in s 355-20 to mean "core R&D activities" or "supporting R&D activities". 11 The term "core R&D activities" is defined in s 355-25. The elements of sub-section (1) in particular are important - 355-25 Core R&D activities (1) Core R&D activities are experimental activities: (a) whose outcome cannot be known or determined in advance on the basis of current knowledge, information or experience, but can only be determined by applying a systematic progression of work that: (i) is based on principles of established science; and (ii) proceeds from hypothesis to experiment, observation and evaluation, and leads to logical conclusions; and (b) that are conducted for the purpose of generating new knowledge (including new knowledge in the form of new or improved materials, products, devices, processes or services). (2) However, none of the following activities are core R&D activities: (a) market research, market testing or market development, or sales promotion (including consumer surveys); (b) prospecting, exploring or drilling for minerals or *petroleum for the purposes of one or more of the following: (i) discovering deposits; (ii) determining more precisely the location of deposits; (iii) determining the size or quality of deposits; (c) management studies or efficiency surveys; (d) research in social sciences, arts or humanities; (e) commercial, legal and administrative aspects of patenting, licensing or other activities; (f) activities associated with complying with statutory requirements or standards, including one or more of the following: (i) maintaining national standards; (ii) calibrating secondary standards; (iii) routine testing and analysis of materials, components, products, processes, soils, atmospheres and other things; (g) any activity related to the reproduction of a commercial product or process: (i) by a physical examination of an existing system; or (ii) from plans, blueprints, detailed specifications or publically available information; (h) developing, modifying or customising computer software for the dominant purpose of use by any of the following entities for their internal administration (including the internal administration of their business functions): (i) the entity (the developer) for which the software is developed, modified or customised; (ii) an entity *connected with the developer; (iii) an *affiliate of the developer, or an entity of which the developer is an affiliate. 12 The term "supporting R&D activities" is defined by s 355-30 - 355-30 Supporting R&D activities (1) Supporting R&D activities are activities directly related to *core R&D activities. (2) However, if an activity: (a) is an activity referred to in subsection 355-25(2); or (b) produces goods or services; or (c) is directly related to producing goods or services; the activity is a supporting R&D activity only if it is undertaken for the dominant purpose of supporting *core R&D activities. 13 Subdivision 355-D deals with notional deductions for R&D expenditure. In broad terms, the effect of this Subdivision is that an R&D entity can notionally deduct its expenditure on registered R&D activities for which certain conditions are met.

The IR&D Act 14 Section 4 of the IR&D Act provides that the terms "core R&D activities", and "supporting R&D activities" have the same meaning as in the ITAA 1997. 15 Registration of an R&D entity for R&D activities is dealt with in Pt III of the IR&D Act. Under that Part (in particular, Division 2 of that Part) - (1) the Board may register an R&D entity for R&D activities conducted during an income year; (2) the Board may make findings about the nature of an R&D entity's activities, both before and after registration; and (3) such findings bind the Commissioner for the purposes of any entitlement of the R&D entity to a tax offset under Division 355 of the ITAA 1997 for the activities. 16 Section 27A of the IR&D Act deals with registering R&D entities for R&D activities. Sub-section 27A(3) provides in relation to the registration of supporting R&D activities - (3) For each activity registered under subsection (1) as a supporting R&D activity for an R&D entity for an income year, the registration is to also specify: (a) one or more activities as the corresponding core R&D activities; and (b) if any of those activities specified as a core R&D activity is not registered under paragraph (1)(a) for the R&D entity for the income year - each income year for which that core R&D activity: (i) was registered under paragraph (1)(a) for the R&D entity; or (ii) is proposed to be registered under paragraph (1)(a) for the R&D entity. 17 Sub-section 27F(1) provides that the Board may conduct one or more examinations of all or part of an R&D entity's registration under s 27A for an income year for the purposes of making one or more findings under s 27J(1) (set out below). The note under s 27F(1) explains that a finding under s 27J(1) will support the entity's registration, or cause the variation of that registration. 18 The key provision of the IR&D Act for present purposes is s 27J - Findings about a registration (1) The Board may make one or more findings to the following effect about an R&D entity's registration under section 27A for an income year (the registration year): (a) that all or part of a registered activity was a core R&D activity conducted during the registration year; (b) that all or part of a registered activity was not an activity of a kind covered by paragraph (a); (c) that all or part of a registered activity was a supporting R&D activity conducted during the registration year and in relation to: (i) one or more specified registered core R&D activities; or (ii) one or more specified core R&D activities for which the entity has been registered in an earlier income year; or (iii) one or more specified core R&D activities yet to be conducted for which the entity could be registered in the registration year if those activities were conducted during the registration year; or (iv) several specified core R&D activities, each covered by subparagraph (i), (ii) or (iii); (d) that all or part of a registered activity was not an activity of a kind covered by paragraph (c). Note 1: A finding is reviewable (see Division 5). Note 2: The Board could make a finding under paragraph (b) if, for example, the Board has insufficient information to make a finding under paragraph (a). Similarly, the Board could make a finding under paragraph (d) if it has insufficient information to make a finding under paragraph (c). (2) If the Board makes a finding under subsection (1) in relation to the R&D entity's registration, the Board may specify in the finding the times to which the finding relates. Example: A finding under paragraph (1)(a) could specify the times during the registration year that a registered activity was a core R&D activity. (3) This section has effect subject to section 32B (findings cannot be inconsistent with any earlier findings). 19 The effect of s 27J is that the Board may make findings as described in paragraphs (a), (b), (c), or (d) of sub-section (1) in respect of registered activities. For example, if an activity is registered for a particular income year, the Board may make a finding pursuant to paragraph (a) that the activity was a "core R&D activity" and that it was conducted during the income year. By way of further example, if an activity is registered for a particular income year, the Board may make a finding pursuant to sub-paragraph (c)(ii) that the activity was a "supporting R&D activity" conducted in the income year, in relation to one or more specified "core R&D activities" for which the entity was registered in an earlier income year. In addition, under s 27J(1)(c)(iii), the Board could make a finding that all or part of a registered activity was a supporting R&D activity conducted during the registration year in relation to "one or more specified core R&D activities yet to be conducted for which the entity could be registered in the registration year if those activities were conducted during the registration year". If activities were registered as "core R&D activities", the Board could nevertheless find that they were "supporting R&D activities" pursuant to paragraph (c). 20 Section 27K deals with notice of the Board's findings, or of decisions refusing to make findings. Section 27L deals with automatic variations so that an R&D entity's registration is consistent with the Board's findings. 21 Division 5 of Pt III of the IR&D Act deals with the review of decisions of the Board. Section 30D provides for a process of internal review, and s 30E provides for external review by the Tribunal of internal review decisions.

The applicant's applications for registration 22 The applicant lodged applications for the registration of R&D Activities for each of 2014, 2015, and 2016. In circumstances that I will explain later, only the respondent's reviewable decisions in relation to the applications for the 2014 and 2015 years were before the Tribunal in the proceeding the subject of this appeal.

2014 year 23 In relation to the 2014 year, the applicant's application to the respondent listed its R&D projects and activities as follows - 1 - UVI001 Design and Development of an integrated Health and Fitness program and Cloud based Decision Support Systems 1.1 Design of Fitness management algorithms for calorie consumption measurement 1.1.1 - Supporting 1.2 - Design of Health management algorithms for calorie intake measurement 1.2.1 - Supporting 1.3 - Conceptual design and evaluation of a potential implementation of Cloud based Decision Support Systems 1.3.1 - Supporting 24 Activity 1.1 which the applicant claimed it had undertaken as a core activity in the 2014 year, being the design of the fitness management algorithms for calorie consumption measurement, was described by the applicant in its application for registration as follows - HYPOTHESIS Create a set of Fitness Algorithms (decision software driver modules) capable to personalize a predefined Fitness Program in accordance with the fitness objectives and personal health information of the end user, based on a set of generic rules to be determined - our hypothesis is that the consumption of calories for the same amount of Fitness exercises is dependent upon the age, activity habits and health condition of each individual. EXPERIMENTS The experiments targeted groups of individuals of different ages, sex, personal habits and health status, who performed for a predetermined period of time the same type of physical exercises confined to the same number of repeats, w[e]ights and time of the day. Experiment 1 - Determine the consumption of calories for a set of pseudo-fix parameters such as Inactive people, active people, gym fanatics, same sex, different sex and same age participants, different age groups, overweight and normal weight. Experiment 2 - Determine consumption of calories for an (sic) set of variable parameters such as time of day for the exercises, relaxed and under stress moods, healthy and sick medical status, and number of repeats for the experiments with the same group type. OBSERVATION OF THE RESULTS The outcome of the tests provided us with the basic set of criteria which will be used in the development of Fitness algorithms: • A static person consumes more calories than an active person for the same amount of physical activity; • An elderly person consumes more calories for the same amount of physical activity; • A sick person consume[s] more calories than a healthy person for the same amount of physical activity. CONCLUSION The unknown element of this core activity was the process of combining the personal variables listed in Experiment 1 with the time and mental status of the individuals tested with Experiment 2. These results combined with the personal set of Fitness Objectives enabled us to move from empirical observation to mathematical model[l]ing of calorie consumption for each age and sex group and make use of existing fitness measurement devices, programs and libraries containing statistical information. 25 The supporting activity 1.1.1 that the applicant claimed was carried out in the 2014 year was described in its application for registration as follows - In order to test our hypothesis that the consumption of calories is dependent upon the personal parameters of each individual we researched an extensive number of Fitness and Medical magazines. In addition, we have reviewed the functional specifications of the latest fitness data capture devices (i.e. Android mobile phones, FitBit smart watches, Up 24 trackers, etc), fitness programs (GOOGLE, SAMSUNG, etc) and Industry problems (potential health hazards and physical body damage). Our preparation for testing included but was not limited to designing a set of PC based generic interfaces for devices which record the calorie consumption for various activities (i.e. walking, swimming, rowing, gym, etc) and a set of models to determine the outcome of various Fitness activities in relation to the personal health status and personal fitness objectives. - Define the set of pseudo-fix personal parameters - Define the set of variable program parameters - Define a set of most likely Fitness objectives - Establish sets of Fitness activities for each Fitness objective - Create a set of tables for calories consumption taking into consideration the parameters listed herein above This comprehensive set of references has created a[n] intelligence framework for developing the Fitness Algorithms targeted by our hypothesis. 26 Activity 1.2 which the applicant claimed it had undertaken as a core activity in the 2014 year, being the design of health management algorithms for calorie intake measurement, was described by the applicant in its application for registration as follows - HYPOTHESIS Create a set of Health Algorithms (decision software driver modules) capable to personalize a predefined Health Program in accordance with the health objectives and personal health information of the end user, based on a set of generic rules to be determined - our hypothesis is that the retention of calories from the same amount of food and beverages is dependent upon the age, activity habits and health condition of each individual. EXPERIMENTS Our objective was to determine a set of generic rules which taking into consideration the personal details and the personal fitness goals will personalize the Dietary program accordingly by performing experiments targeting groups of individuals of different ages, sex, personal habits and health status, who will consume for a predetermined period of time the same amount and types of foods and beverages at the same time of the day. Experiment 1 - Determine the intake of calories for a pseudo-fix set of personal parameters - inactive people, active people, health fanatics, same sex and different sex individuals, same age and different age groups, overweight and normal weight. Experiment 2 - Determine the intake of calories for a set of variable parameters - time of day for meals, relaxed and stress moods, healthy and sick medical status and number of experiment repeats with the same groups. OBSERVATION OF RESULTS The outcome of the tests provided us with a basic set of criteria which will be used in the development of Health algorithms. • For a static person the intake of calories is greater than an active person from the same amount of food and beverages; • For an elderly person the intake of calories is less than for a young person for the same amount of food and beverages; • For a sick person the intake of calories is less than for a healthy person; ETC CONCLUSION The unknown element of this core activity was the process of combining the personal variables listed in Experiment 1 with the time and mental status of the individuals tested with Experiment 2. These results combined with the personal set of Health Objectives enabled us to move from empirical observation to mathematical modelling of calorie intake for each age and sex group and make use of existing calorie measurement devices (i.e. food, beverages, medicines), programs and libraries containing statistical information. 27 The supporting activity 1.2.1 that the applicant claimed was carried out in the 2014 year was described in its application for registration as follows - In order to test our hypothesis that the intake of calories is dependent upon the personal parameters of each individual we researched an extensive number of Medical magazines and Dietary programs. In addition, we have reviewed the functional specifications of existing calorie measurement and capture devices (i.e. food scanners, food scales, heart bit monitors, blood pressure monitors, etc), calorie content of prescribed drugs and likely industry problems (potential health hazards). Our preparation for testing included but was not limited to designing a set of PC based generic interfaces for devices which record the calorie intake for various calories sources (i.e. foods, beverages and drugs) and a set of models to determine the outcome of various Directory programs in relation to the personal Health status and personal Health objectives. - Define the set of pseudo-fix personal parameters - Define the set of variable program parameters - Define a set of most likely Health objectives - Establish sets of dietary programs for each Health objective - Create a set of tables for calories contents taking into consideration the parameters listed herein above This comprehensive set of references has created a[n] intelligence framework for developing the Health control Algorithms targeted by our hypothesis. 28 Activity 1.3 which the applicant claimed it had undertaken as a core activity in the 2014 year, being the conceptual design and evaluation of a potential implementation of cloud based decision support systems, was described by the applicant in its application for registration as follows - HYPOTHESIS Create a "well being" Decision Support System program, Cloud based, which combines 3rd party Health and Fitness programs with personal Health and Fitness objectives with the personal Medical and Fitness status, to ensure a correct interpretation of collision possibilities between such programs and personal objectives. Since such an integrated program does not currently exist we should build one based on the models resulting from the Core activities 1 and 2 above and the set of decision rules determine[d] by a new set of experiments. EXPERIMENTS A set of experiments was defined to determine the validity of our hypothesis and of the proposed decision making process. This has required the careful consideration of certified health programs made available by medical practitioners, the availability of personal details and health status data and the existence of fitness programs developed and made available by professional fitness specialists. Experiment 1: Design and implement a Cloud based Fitness monitoring program to captures (sic) the results of various Fitness exercises undertaken by the user in accordance with Fitness programs certified by reputable organizations or individual trainers. Experiment 2: Design and implement a Cloud based Dietary monitoring program which captures the results of Food, Beverages and Pharmaceutical products intake by the user in accordance with personal Dietary programs certified by reputable organizations or individual diet experts. Experiment 3: Design and implement a Cloud based Decision Support System which will process the inputs received from the Fitness and Dietary programs and of the personal Health and Fitness status and objectives for the purpose of ensuring that the individual's health is not damaged or endangered by either the personal Fitness program of (sic) by the personal Dietary program and that the personal Health and Fitness objectives are achieved. OBSERVATION OF RESULTS Whilst performing the tests it was found that the dynamic tailoring of predefined Health and Fitness programs in accordance with the personal Health and Fitness status and the personal Health and Fitness objectives did not reach our desired accuracy of 100% (as required by the Health Industry) and therefore more mathematical modelling is required. The main obstacle encountered by our experiments was the complexity of mathematical models which required strict observance of the 4 input sets of parameters for the production of one health safe Health & Fitness program. CONCLUSION It was found that there are no statistics or comparable models for reference and therefore the outcome of these Cloud based Health and Fitness algorithms and programs is unknown and therefore it represents the greatest risk for us as developers since any errors may endanger the health or even the life of its users. 29 The supporting activity 1.3.1 that the applicant claimed was carried out in the 2014 year was described in its application for registration as follows - In order to test our hypothesis we have researched a number of typical Cloud based applications, infrastructure services and storage facilities which can be used in conjunction with the proposed Health and Fitness programs (i.e. APPLE, Samsung, Google, etc). The prevention of health hazards was the underlying element of the development of Cloud based Decision Support Systems, accessible by mobile phone users, which MUST ensure compliance between the 6 types of programs (generic health and fitness programs and personal health fitness status and personal health and fitness objectives). A set of intelligence gathering models have been designed to support the decision support system which combines the Core activities 1 and 2 herein before described. We plan to attract Health and Fitness experts to evaluate the results of our Health and Fitness mathematical models and assist us in offering a reliable Health and Fitness Application for mobile phone users. Further, we will seek acceptance by accredited medical organizations, government agencies and health insurance companies which will also be beneficiaries of a healthy population (i.e. reduced health bills at national level and lower insurance risks).

2015 year 30 In relation to the 2015 year, the applicant's application to the respondent listed the R&D Projects and Activities as follows - 1 - UVI001 Design and Development of an integrated Health and Fitness program and Cloud based Decision Support Systems 1.1 - Improvements of Fitness monitoring algorithms and Multi level Search Engine 1.1.1 - Supporting 1.2 - Improvements of Diet monitoring algorithms and Multi level Search engine 1.2.1 - Supporting 1.3 - Improvements of Cloud based anti-collision systems with smoothing algorithms 1.3.1 - Supporting 31 Activity 1.1 which the applicant claimed it had undertaken as a core activity in the 2015 year, being the improvement of the fitness monitoring algorithms and multi-level search engine, was described by the applicant in its application for registration as follows - Our hypothesis was that calorie burning is directly dependent on personal fitness and health status and health history. After a comprehensive number of systematic experiments we have found a number of limitations and inaccuracies in our hypothesis. We had to experiment how a set of apparently unrelated factors (i.e. elevation, temperature, humidity, etc) would affect the calorie burning rate and how could the accuracy and speed of search for suitable fitness programs can (sic) be improved. The outcome of the experiments demonstrated that our original hypothesis was limited and that we have to include in our program search engines and calorie burning calculation algorithms these variables. The logical conclusion was that we have to modify our fitness program search engine and calorie burning calculation algorithms within the context of user's fitness objectives, health and fitness status and health history. 32 The supporting activity 1.1.1 that the applicant claimed was carried out in the 2015 year was described in its application for registration as follows - In order to test our hypothesis that the consumption of calories is dependent upon the personal parameters of each individual we researched an extensive number of Fitness and Medical magazines, as part of the preliminary preparation of the experimental phase, we have developed a comprehensive testing plan of the experiments to determine the outcome of various fitness activities in relation to the personal health history, health status and personal fitness objectives of the individual. Determine the accuracy and efficiency of the fitness search engine and auto-correction fitness program: - Test the effects of pseudo-fix personal parameters (age, height, weight, gender, fat %) - Test the effects of variable program parameters (blood pressure, cholesterol, sugar level, activity) - Test the limitations imposed by the Fitness objectives (reduce weight, increase strength, etc) - Identify and retain sets of Fitness activities for each Fitness objective - Establish links to tables listing calorie burning rates for various fitness exercises taking into consideration the parameters listed above This comprehensive set of experiments has resulted in the creation of an intelligence framework which enhances the accuracy and processing speed of the Fitness Algorithm and search engine supporting our revised hypothesis. 33 Core activity 1.2 which the applicant claimed it had undertaken in the 2015 year, being the improvements of diet monitoring algorithms and multi-level search engine, was described by the applicant in its application for registration as follows - Our hypothesis was that the calorie intake is directly dependent on personal fitness and health status and health history. After a comprehensive number of systematic experiments we have found a number of limitations and inaccuracies in our hypothesis. We had to experiment (sic) [demonstrate] how a set of apparently unrelated variables (i.e. elevation, temperature, stress etc) would affect the calorie intake rate and how could the accuracy and speed of search for suitable health programs can (sic) be improved. The outcome of the experiments demonstrated that our original hypothesis was limited and that we have to include in our program search engines and calorie intake calculation algorithms the new variables. The logical conclusion was that we have to modify our health program search engine and calorie intake calculation algorithms with the context of the user's health objectives, health status and health history to obtain a greater degree of accuracy. 34 The supporting activity 1.2.1 that the applicant claimed was carried out in the 2015 year was described in its application for registration as follows - In order to test our hypothesis that intake of calories is dependent upon the personal parameters of each individual we researched an extensive number of Fitness and Medical magazines, as part of the preliminary preparation of the experimental phase, we have developed a comprehensive testing plan of the experiments to determine the outcome of various dietary programs in relation to the personal health history, health status and personal health objectives of the individual. Determine the accuracy and efficiency of the health search engine and auto-correction health program: - Test the effects of pseudo-fix personal parameters (age, height, weight, gender, fat %) - Test the effects of variable program parameters (blood pressure, cholesterol, sugar level, activity) - Test the limitations imposed by the Health objectives (reduce blood pressure, reduce sugar level, reduce cholesterol, etc) - Expand the setoff variable program parameters (temperature, altitude, stress level, time of day) - Identify and retain sets of Health diets for each Health objective - Establish links to tables listing calorie contents of various foods and beverages taking into consideration the parameters listed above This comprehensive set of experiments has resulted in the creation of an intelligence framework which enhances the accuracy and processing speed of the Health Algorithm and search engine supporting our revised hypothesis. 35 Core activity 1.3 which the applicant claimed it had undertaken in the 2015 year, being the improvements of cloud based anti-collision systems with "smoothing" algorithms, was described by the applicant in its application for registration as follows - We have determined from our experiments that our Cloud based anti-collision program elimination hypothesis was unsustainable in extreme cases such as the absence of suitable Fitness and/or Health programs due to user's personal health restrictions and personal wellbeing Objectives. To offer end users suitable Health & Fitness programs we have to rethink our strict anti-collision approach and modify the decision support systems logic to accommodate part of, if not all of the end users' Objectives, without compromising their health. A number of "smoothing" algorithms have been appended to the AI search engines and DSS integration programs with the purpose of offering the most suitable diets and fitness exercises which achieve part of, if not all of the user's Objectives. It should be noted that without re-thinking our AI and DSS architectures, the proposed H&F Programs will be singularly relevant to a limited number of end users. The flexibility offered by the proposed "smoothing" algorithms will generalise the outcome and applicability of this project. Whilst performing the tests it was found that the dynamic tailoring of predefined Health and Fitness programs in accordance with the personal Health and Fitness status and the personal Health and Fitness objectives did not reach our desired accuracy of 100% (as required by the Health Industry) and therefore more mathematical modeling (sic) is required. The main obstacle encountered by our experiences was the complexity of mathematical models which required strict observance of the 4 input sets of parameters for the production of one health safe Health & Fitness program. This complexity was further expanded by the addition of the new sets of environmental parameters listed in the Core activities 1 and 2. In terms of knew (sic) knowledge it was found that the addition of environmental factors (variables) affecting calorie intake and burning and the extension of our AI algorithms and DSS engines resulted in faster and more accurate calculation algorithms and the provision of at least one suitable H&F program. The extreme case of no suitable programs should be automatically referred in the future to medical and fitness experts for resolution. 36 The supporting activity 1.3.1 that the applicant claimed was carried out in the 2015 year was described in its application for registration as follows - In order to test our modified hypothesis we have researched a number of typical Cloud based health and fitness programs offered by leading organizations such as APPLE and Google. A comprehensive test plan was developed including experiments deemed as necessary not only to prevent health hazards but improve the AI algorithms and DSS engines in terms of accuracy and speed through the integration of additional factors affecting the calorie intake and burning and the "smoothing" of end user's Objectives. We have tested design test simulation tools to test the accuracy of our AI algorithms and DSS engines for: - Normal conditions (strict observance of health restrictions) - Boundary conditions (employing "smoothing" algorithms), and - Transient conditions (dynamic changes on the health and fitness status) The outcome of our experiences helped us to clearly define the end users' choices: discard programs which do not fit, or revise and modify personal objectives, or tailor recommended programs to fit personal objectives and to build the support framework needed by this Core activity.

2016 year 37 In relation to the 2016 year, in its application for registration the applicant identified two core and two corresponding support activities as follows - 1-1: Design and development of an integrated Health and Fitness program and Cloud based Decision Support Systems 1.1 Core 1: Development and testing of Cloud based emulations of payment applications using POE, IKM and tokenisation methods. 1.1.1- SUPPORT 1: Development and testing of several mobile payment applications 1.2 Core 2: Development and testing of Cloud based decision support systems and artificial intelligence engines for integrated health and fitness programs 1.2.1- SUPPORT 2: Development and testing of a range of health fitness mobile applications 38 Core activity 1.1 that the applicant claimed was carried out in the 2016 year, being the development and testing of cloud based emulations of payment applications using "POE", "IKM" and tokenisation methods, was described by the applicant in its application for registration as follows - HYPOTHESIS (OF EXPERIMENTS): Our hypothesis was that by tokenising payment transactions, in the same way we do for medical records, we could use non-secure mobile phones to perform local and international electronic fund transfers in a secure manner. EXPERIMENTS: The difficulty was in finding how to authorize and capture payment transactions from "unknown" end-users' devices and transaction processors. Experiment 1: determine how EFT transactions can be performed between the accounts of the same user within the same bank and between different banks; Experiment 2: determine how EFT transactions can be performed between the accounts of two different users within the same bank and between different banks; Experiment 3: determine how EFT transactions can be performed between the accounts of the same user or of two different users internationally; Experiment 4: demonstrate how the overall system reacts to the 3 types of testing domains (normal, boundary and transient). UNKNOWN OUTCOME OF EXPERIMENTS/OBSERVATION: There was no prior known methodology to enable us or other skilled persons to predict the outcome of the experiments in advance since for the first time POE, IKM and TOKENISATION were combined into a unitary CLOUD based Payment System. EVALUATION: It was determined that tokenised payments can be performed via a CLOUD based transaction processor which is known to a local bank such as a Retail Integrator or as a PayPal user or as a SWIFT client. NEW KNOWLEDGE/OUTCOME: The positive outcome of this core activity was the creation of a new CLOUD paradigm (POE and IKM) for local and international Electronic Payments. CONCLUSION: We demonstrated that non-secure devices can perform sensitive tasks such as payments in a secure manner is operating in conjunction with Cloud based token security. 39 The supporting activity 1.1.1 for the 2016 year was described by the applicant in its application for registration as follows - Cloud based International payments (via SWIFT) • Deposits - Prior to any fund transfers the user will be required to transfer from his/her bank the intended transfer amount plus the transfer fee or a larger amount to be used for multiple transfers into GSS trust account. The funds are managed by a "trusted" bank and used for international fund transfers via SWIFT. • Withdrawals - Members may withdraw at any time their funds held by GSS in the trust account, since GSS maintains a real-time account balance for each member. • Balance Enquiries - Members may get at their discretion, at any time, their balances of their accounts held by GSS in the trust account, since GSS maintains a real-time account balance for each member. • Fund Transfers - Members will be able to transfer funds internationally to pre-defined overseas accounts (repetitive-line accounts list) subject to funds availability by requesting GSS CLOUD APP to process their request under the strict security rules mentioned above. Cloud based Local payments (via PayPal) • Requesting Funds (REQUEST) -This function will enable the end-user to get certified value TOKENS from the CLOUD based Service Provider APP (GSS). Several certified TOKENS may coexist. GSS will validate the end-user credentials (ID, PSW and DEVICE) and get the requested amount from the end-user's bank account, in case the "virtual" account balance held by GSS is insufficient. GSS will then send a certified value TOKEN representing the amount and flag the TOKEN as "issued". • Transferring Funds (TRANSFER) - APP users will be able to transfer funds (value TOKENS) to other APP users who will accept the funds and provide the goods or services or cash upon receiving confirmation from the GSS Service Provider that the value TOKEN certificate is valid. RECIPIENT DEVICES are smart-phones, EFTPOS devices and ATMs. GSS will validate the value TOKEN certificated and in accordance with AMLCTA act, will save the transfer trace for future audit. 40 Core activity 1.2 that the applicant claimed was carried out in the 2016 year, being the development and testing of cloud based decision support systems and artificial intelligence engines for integrated health and fitness programs, was described by the applicant in its application for registration as follows - HYPOTHESIS (OF EXPERIMENTS): Our hypothesis was that we could integrate 3 generic dietary, fitness and drug prescription programs with the personal health and fitness status within the context of personal health and fitness objectives in an interactive manner within strict safety and security regulations. EXPERIMENTS: The challenge was how such a complex framework of parameters can be integrated into a safe personal well-being program in order to develop the decision support engines required to select the most adequate health and fitness programs and build the artificial intelligence engines required to eliminate collision of such programs and adjust the end-user's goals and objectives within pre-determined safety boundaries. • Experiment 1: determine how a fitness program is impacted by the other 5 programs. • Experiment 2: determine how a dietary program is impacted by the other 5 programs. • Experiment 3: determine how a medical program is impacted by the other 5 programs. • Experiment 4: demonstrate how the overall system reacts to the 3 types of testing domains (normal, boundary and transient). UNKNOWN OUTCOME OF EXPERIMENTS/OBSERVATION: After in-depth research we could not find any system or known methodology to enable us or other skilled persons to predict the outcome of the experiments in advance since for the first time DIETARY and FITNESS programs were combined with the personal HEALTH and FITNESS STATUS, OBJECTIVES and PROFILE of the end-user into a unitary CLOUD based Health and Fitness system. EVALUATION: It was determined that the interaction of the generic and personal programs can be merged into a satisfactory level of safety through the implementation of "smoothing" of the anti-collision algorithms. NEW KNOWLEDGE/OUTCOME: The positive outcome of our experiments was the creation of a CLOUD based paradigm (H&F) for integrated Health and Fitness programs. CONCLUSION: We demonstrated that it was possible to integrate the 6 types of programs and resolve the safety and security problems encountered by specialised health and fitness programs available at the time on the market. 41 The supporting activity 1.2.1 for the 2016 year was described by the applicant in its application for registration as follows - Test Plan considerations: The variable nature of the amount of calories received from food and beverages intake and the amount of calories spent as a result of various fitness exercises must be combined with the personal information of the individual (Profile and H&F Status). Since we have introduced another variable in the form of Medical Prescriptions monitoring we have to rethink the Test Plan of the DSS and A1 engines in order to ensure compatibility across the entire spectrum of H&F algorithms. R&D focus - Fitness DSS algorithm The DSS Fitness algorithm identifies the most suitable Fitness programs which fit the end-user's Fitness Objectives within the limitations of the end-user's Personal Profile and H&F Status. R&D focus - Diet DSS algorithm The DSS Diet algorithm identifies the most suitable Dietary programs which fit the end-user's Health Objectives within the limitations of the end-user's Personal Profile and H&F Status. R&D focus - Anti-collision A1 engine The Anti-Collision A1 engine ensures that the Fitness and Diet programs, when integrated, do not exclude each other within the context of end-user's Personal Profile and H&F Status R&D focus - H&F program smoothing A1 Engine The Smoothing A1 engine ensures that the end-user's Objectives are revised when all the selected fitness and Diet programs have been excluded by the Anti-Collision A1 engine in order to enable the end-user to participate to an H&F program within realistic Objectives.

The respondent's findings 42 On 3 November 2016, the respondent notified the applicant of findings that had been made in relation to the 2014 and 2015 years, namely that the applicant's registered activities for those years were ineligible as core R&D activities, or as supporting R&D activities. The applicant then sought an internal review of the findings. 43 On 3 March 2017, the respondent advised the applicant of the result of the internal review, which was to confirm the decision that the applicant's registered activities for 2014 and 2015 were not eligible. On the same day, the respondent also notified the applicant of findings made at first instance in relation to the applicant's registered activities for the 2016 year, namely that those activities also were not eligible. The respondent's decision upon its internal review in respect of the 2014 and 2015 years was reviewable by the Tribunal at this stage, but not the findings in relation to the 2016 year because at that stage there was no reviewable decision following internal review: IR&D Act, s 30A, s 30E.

The application to the Tribunal 44 On 9 March 2017, the applicant lodged an application with the Tribunal seeking review of the decision of the respondent upon internal review relating to the applicant's registered activities for 2014 and 2015. The application also referred to the 2016 year and attached a copy of the respondent's certificate dated 3 March 2017 containing its findings in respect of the 2016 year, notwithstanding that at that point in time the respondent had not made a reviewable decision in relation to the 2016 year. 45 Subsequently, on 25 April 2017, the applicant sought an internal review of the respondent's findings in relation to the 2016 year.

Preliminary steps before the Tribunal 46 There was evidence before the court about the progress of the applicant's applications before the Tribunal, and about two telephone directions hearings before the Tribunal conducted on 14 September 2018 and 23 October 2018. This evidence was tendered with leave, and was relevant to those grounds of appeal that complained that the Tribunal had erred in failing to consider the application for review in respect of the 2014 and 2015 years together with an application for review in respect of the 2016 year. The evidence on these matters was extensive. It included affidavits of Mr Werner Nicolau, the sole director of the applicant, and his father Mr Mark Nicolau about what occurred at the directions hearings, in circumstances where there was no transcript available to the court. The evidence also included an affidavit of the respondent's solicitor sworn 26 October 2020 which comprised 456 pages, including its 60 annexures. 47 In relation to the review before the Tribunal, there were communications between the solicitors for the respondent and the solicitors then acting for the applicant, Halperin & Co. In a letter from the respondent's solicitors to the applicant's solicitors dated 3 May 2017, it was specifically put to the applicant's solicitor that he had advised that the applicant would not seek to stay the proceedings before the Tribunal pending the outcome of the respondent's internal review in respect of the 2016 year. Draft timetabling orders were proposed by the respondent consistently with this position. 48 On 30 June 2017, new solicitors for the applicant, Portfolio Law, filed a statement of issues with the Tribunal. The reviewable decision that was referred to at [10] of that statement was the respondent's decision of 3 March 2017 to uphold the respondent's decision of 3 November 2016, which was in respect of the 2014 and 2015 years. 49 There is evidence that on 31 October 2017, the respondent notified the applicant of the result of its internal review for the 2016 year, which was to confirm the decision. The evidence is in the form of a copy of an email produced by the respondent dated 31 October 2017 to Mr Werner Nicolau, which attached the respondent's decision upon internal review for the 2016 year. The copy email is accompanied by a read receipt from what appears to be the email address of Mr Werner Nicolau, which would support an inference that the email was read on 31 October 2017 at 9.26 pm. However, it is a curiosity that the read time is precisely the same as the send time, down to the second. The way the read receipt was generated, and its significance, were not explained by the evidence. As I indicate below, there was a dispute on the evidence as to whether that notification was received by the applicant. 50 On about 16 November 2017, the applicant filed and served a statement of facts issues and contentions before the Tribunal. That statement at [2] referred only to the 2014 and 2015 years as being the relevant income years in issue. 51 On about 1 February 2018, the Tribunal listed the matter for hearing on 16 and 17 May 2018. 52 On 7 March 2018, the respondent filed and served its statement of facts issues and contentions before the Tribunal. The respondent referred to the applicant's application for internal review of the decision in respect of the 2016 year, but not to the fact that the review had been determined, or to the notification dated 31 October 2017. However, the statement squarely advanced the contention that the respondent's application to the Tribunal did not extend to a reviewable decision in respect of the 2016 year. 53 There were delays by the applicant in filing evidence in reply, and on 26 April 2018, its solicitors withdrew. At a directions hearing on 27 April 2018 the Tribunal directed the applicant to file its reply evidence by 4 May 2018, and confirmed the hearing dates, which were then 16 and 17 May 2018. At a further directions hearing on 11 May 2018, the time for the applicant to file reply evidence was further extended to 16 May 2018, and the May hearing dates were vacated. 54 On about 17 May 2018, the applicant, which by now was not legally represented, filed and served a statement of facts issues and contentions in reply. 55 A further directions hearing was conducted by the Tribunal on 19 June 2018 at which the matter was referred to a listing clerk. On about 1 August 2018, the Tribunal issued notices advising that the proceeding was listed for hearing on 29 and 30 October 2018. 56 On 7 August 2018, the applicant filed and served a brief further witness statement of Mr Werner Nicolau. The material substance of that statement was that Mr Nicolau had not received a response from the respondent to its request to review its decision in respect of the 2016 year. 57 In response, on 20 August 2018 the respondent's solicitors wrote to the applicant and to the Tribunal drawing attention to the respondent's decision on 31 October 2017, to which I referred at [49] above. The letter attached a copy of the internal review decision for the 2016 year, the covering letter to the applicant, the email attaching those documents and a copy of the read receipt from Mr Nicolau. The respondent's solicitors foreshadowed that the respondent would resist any attempt to bring the reviewable decision for the 2016 year into the proceeding on the basis that it may require additional evidence and a further delay in the proceedings. The solicitors for the respondent invited the applicant to approach the Tribunal for directions if it intended to press the Tribunal to consider the 2016 year. 58 Mr Werner Nicolau stated in an affidavit before this court affirmed 8 December 2020 that he did not receive the respondent's reviewable decision dated 31 October 2017 in relation to the 2016 year until the respondent emailed a copy on 20 August 2018. Mr Nicolau was not cross-examined on this evidence, and to the extent that it is relevant to this appeal I accept that Mr Nicolau was not aware until 20 August 2018 that the respondent had made reviewable findings in relation to the 2016 year. Mr Nicolau's direct affidavit evidence on this point is supported by the surrounding circumstantial evidence. If Mr Nicolau had become aware of the findings for the 2016 year when they were apparently sent to him by the respondent, it is likely that he would have taken steps earlier than he did to apply to have the findings reviewed. The issues on appeal do not call for any findings to be made about whether the respondent succeeded in sending the findings for the 2016 year to Mr Nicolau by email as its business records would indicate, and if the email had been sent on 31 October 2017, the reasons why Mr Nicolau did not become aware of the email at about that time. 59 In the applicant's reply submissions to the court the applicant claimed that the respondent's statement of facts issues and contentions before the Tribunal dated 7 March 2018 had omitted to state that that a reviewable decision had been made in respect of the 2016 year, as a result of which the applicant remained unaware of the decision until 20 August 2018 which led to the application at the directions hearing on 14 September 2018. The evidence does not support any inference that the respondent consciously omitted in its statement of 7 March 2018 to draw the applicant's attention to its reviewable decision of 31 October 2017 in respect of the 2016 year. This is for three reasons. First, the business records of the respondent in the form of the copy email dated 31 October 2017 would have indicated to the respondent that the decision was sent to the applicant. Second, there is no inference to be drawn that the respondent had any knowledge prior to 20 August 2018 that the applicant was unaware of the decision. The third reason is that when it became apparent in August 2018 that the applicant was unaware of the decision, the respondent's solicitors provided a copy of the decision to the applicant: see [57] above. 60 At a telephone directions hearing before the Tribunal conducted on 14 September 2018, the applicant sought to have the review of the decisions for the 2014 and 2015 years heard together with a review of the respondent's decision in respect of the 2016 year. Mr Mark Nicolau set out in his affidavit affirmed 8 December 2020 his account of what occurred at the directions hearing on 14 September 2018 - 9. At the hearing, amongst other things, I recall that I asked the Tribunal to include the review of the Respondent's decision in respect of the 2015/2016 year. I made submissions to the Tribunal setting out 6 points as to why the Tribunal should include the Respondent's 2015/2016 year's decision in its review. I made submissions to the Tribunal based upon notes that I had with me, which stated as follows: "Why Should the Tribunal include an external review of the Respondent's R&D decision in the 2015/16 year? 1. BECAUSE the UVI Application for Review to the Tribunal on 9 March 2017 was in respect of three years that is the RD incentives for the years ended 30 June 2014 30 June 2015 and 30 June 2016. 2. BECAUSE: the R&D decisions in each year that are the subject of our Application for review arise out of the RD activities into the development of the Health and Fitness System Application. 3. BECAUSE: UVI's complaints and application to review in respect of each year arise from the same facts relating to the conduct and decisions of the Respondent. Each of the years raise common questions of law or fact. 4. BECAUSE: reviewing all three years is necessary to ensure that all questions in the proceeding are effectively and completely determined and decided upon 5. BECAUSE: it is just and convenient and in the interests of justice to determine the issues in respect of all three years in one proceeding rather than having UVI commence separate proceeding which may not even be possible in respect of 2015/16. 6. We did not know until about three weeks ago on 20 August 2018 that a review decision had been made in the 2015/16 year. That does not change the fact that our Application for Review to the Tribunal sought a review of the RD decision for the 2015/16 year. 10. I recall that Counsel for the Respondent said that the Respondent opposed the inclusion of the decision in the 20152016 year in the review. 11. The Tribunal agreed with the Respondents position (Emphasis in original.) 61 Mr Nicolau's account was not challenged, and is not inconsistent with the account of the solicitor for the respondent, Ms Rebekha Pattison, in her affidavit of 26 October 2020 - 73 The parties appeared for a directions hearing on 14 September 2018 before Member Maryniak QC. Ms Melanie Baker of Counsel appeared for the Respondent, instructed by myself. The Member decided that the Applicant would not be allowed to rely on the evidence of Mr Surdut in the absence of a witness statement. The parties then agreed that the Applicant could rely upon the Third Mark Nicolau Statement despite the late service. 74 The parties also discussed the 2015/16 income year during that directions hearing. The Member decided that the 2015/16 income year was not a part of the Application for Review before the Tribunal. 62 I find that at the 14 September 2018 directions hearing the Tribunal determined that any review of the respondent's findings for the 2016 year would not be considered as part of the extant application for review that the Tribunal had fixed for the hearing, and which was due to commence on 29 October 2018. I find that the Tribunal came to the same conclusion at a further telephone directions hearing on 23 October 2018. That finding is supported by the direct evidence of Ms Pattison in her affidavit - 76. The parties appeared for a further directions hearing by telephone before Member Maryniak QC on 23 October 2018. Ms Melanie Baker of Counsel appeared for the Respondent, instructed by myself. Member Maryniak QC again decided that the 2015/16 income year would not be addressed in the upcoming hearing. 63 Ms Pattison's evidence is corroborated by an email to the Tribunal dated 23 October 2018 from Mr Mark Nicolau which stated - Today at the hearing we have been informed by Member Marinyak that our application to include the review of AusIndustry's decision alongside the reviews of AusIndustry's decision for years 2013/2014 and 2014/2015 is rejected since it was lodged too late and that the Respondent is opposing such inclusion in our action (2017/1319) listed for trial on 29 and 30 October 2018. No date was given to us to lodge this application for review. We have been forced to represent ourselves [due] to financial constraints. We get limited pro-bono assistance and the finalisation and lodgement of our request for review was delayed due to the busy schedule of our pro-bono lawyer. We respectfully ask AAT to consider the attached application signed by the principal of the Applicant for a review as a new application and list it for hearing in due course. 64 With this email, the applicant submitted out of time an application to the Tribunal to review the findings of the respondent upon the internal review relating to the 2016 year. Subsequently, the applicant applied for an extension of time to commence the application, which the respondent did not oppose, and on 3 January 2019 the extension was granted by a Deputy President of the Tribunal. 65 I also record here that the affidavit of Mr Werner Nicolau affirmed 8 December 2020 that was before the court annexed almost 2,000 pages of documents relating to evidence that the applicant filed with the Tribunal in its separate application to the Tribunal for review of the respondent's decision relating to the 2016 year. I admitted this evidence on the limited basis that its use was confined to the identification of what evidence might have been available to the applicant to place before the Tribunal but for the alleged errors of law that it raised by its notice of appeal. The affidavit's admission into evidence on that basis was not opposed by the respondent.

The statements of facts issues and contentions 66 I have referred already to the parties' statements of facts issues and contentions that were filed ahead of the hearing before the Tribunal. The applicant's statement of facts issues and contentions, which was prepared by its then solicitors, was brief, framing the issues in succinct terms, and identifying the conclusions that were contended for. 67 The respondent, by its statement of facts issues and contentions, put in issue whether there was any evidence to establish that the activities that were claimed for the 2014 and 2015 years were conducted, and submitted, amongst other things, that even if they were, there was insufficient material to conclude that the activities were experimental activities whose outcome could not be known or determined in advance on the basis of current knowledge, information or experience so as to engage the definition of "core R&D activities" in s 355-25 of the ITAA 1997. The respondent submitted, inter alia, that the applicant's descriptions of claimed supporting R&D activities were too vague to allow the Tribunal to understand the precise nature of the activities claimed to have been undertaken, and further that if the applicant did not demonstrate activities that engaged the definition of "core R&D activities", then there could not be any associated "supporting R&D activities" as defined in s 355-30(1) of the ITAA 1997. In the alternative, the respondent submitted to the Tribunal that there was insufficient evidence that the dominant purpose of the supporting R&D activities was to support the core R&D activities. 68 The applicant's statement of facts issues and contentions in reply was prepared by the applicant itself. Many of the contentions appeared to critique the respondent's processes in arriving at the decision under review, and alleged errors in the respondent's reviewable decision as if on an application for judicial review under the Administrative Decisions (Judicial Review) Act 1977 (Cth) (ADJR Act) claiming (inter alia) that the respondent had "breached principles of administrative law", took irrelevant considerations into account, failed to take account of relevant considerations, failed to act in good faith, failed to act without bias, and acted contrary to law. To the extent that there was reference to evidence, it was generally high-level, such as "the evidence provided by the Applicant of thousands of pages of supporting documents substantiate the research and development carried out, including in great detail the tasks and steps in respect of each Activity". 69 The applicant's statement of facts issues and contentions in reply also purported to bring the 2016 year into the Tribunal's review, referring to each of the 2014, 2015, and 2016 years as the "relevant years". Paragraph [25] of the reply statement claimed that the respondent had made findings in respect of the 2016 year "arbitrarily without a proper basis", referring to the respondent's findings in respect of the 2016 year issued on 3 March 2017. The applicant also claimed in its reply statement to the Tribunal at [25] - … (c) Although the Applicant's research and development activities were different in the 2015/2016 income year from the 2013/2014 income year and the 2014/2015 income year, it appears the Respondent cut and pasted the wording in the certificate of findings as the wording was virtually the same in the 2015/2016 year as it was for the previous year. (d) The Respondent's findings in respect of the 2015/2016 year is probative in relation to the Respondent's process of decision making to make the findings in the 2013/2014 and 2014/2015 years. Accordingly, the Applicant seeks that the Tribunal have regard to the Respondent's 2015/2016 decision when considering the Respondent's findings on the previous years, whether or not the Respondent's 2015/2016 decision will be reviewed by the Tribunal. (e) [T]he Respondent did not act in good faith and did not properly consider the merits of the case in respect of the income year ending 30 June 2016. 70 The heading above [30] of the reply statement claimed, that "[t]he correct or preferable decision on the facts is that the R&D activities of the Applicant in the income years ending 2014, 2015 and 2016 are core R&D Activities and Supporting R&D Activities".

The applicant's written evidence 71 The parties filed statements of evidence with the Tribunal. The applicant filed two witness statements of Mr Werner Nicolau and three witness statements of Mr Mark Nicolau. The first witness statement of Mr Werner Nicolau, which was prepared by the applicant's then solicitors, was accompanied by 330 pages of annexures. The annexures included conceptual diagrams, flow charts, and spreadsheets. A significant number of documents comprised "Test Plan" forms identifying the tester as "N Nicolau", some of which were completed by hand, and others entirely in print. These forms appeared to record things such as diet, exercise, and observed weight loss in the tester. By way of example, one form that was expressed to cover the five week period from 26 May 2014 to 29 June 2014 had written in hand under the heading "test actions" - Eliminate take-away food i.e. Indian, Chinese, Pizza as well as any other processed i.e. Mexican and/or sweets dairy ice cream and cakes. Overall [loss] of weight since start in Nov '13 (135.5 kg) is 20 kg which indicates that the objective of 5 kg per period was unrealistic. Resume proper walking distances i.e. 5-6 km 2 times a day and some exercises. 72 Under the heading "test results" the following was completed - Weight [loss] was at the same pace as before although increased exercises were introduced. 73 Other documents appeared to record what was described as a "personal diet", presumably of the tester. Amongst the annexures was a document of 45 pages described as "User Requirements & Feasibility Study" for "Health Fitness Clubs transaction processing system" prepared by Mr Mark Nicolau. Another document was described as an index of research and development manuals, noted as "Version V5 - 15/11/2017". A document titled "Health & Fitness Systems existing service providers" was produced and identified as "draft V.2". Finally, there was included a copy of a spreadsheet that was described as a "Daily Activities Log" spanning the period to 1 July 2013 to 30 June 2016 and comprising 155 pages. 74 The second witness statement of Mr Werner Nicolau was dated 7 August 2018, to which I have already referred at [56] above. The statement was brief, comprising two pages and two annexures. The substance of the statement was that by email dated 25 April 2017 the applicant had requested the respondent to reconsider its determination in respect of the 2016 year, and that it had not received a response. This statement went to the contested issue of fact to which I referred earlier, concerning whether the applicant ever received the respondent's notification dated 31 October 2017 affirming its decision in respect of the 2016 year on internal review. At the hearing of the review, the Tribunal did not admit this statement into evidence on the ground that it related to the 2016 year which was not before the Tribunal. 75 The first witness statement of Mr Mark Nicolau was also prepared by the applicant's former solicitors. The statement set out Mr Nicolau's background, and described research that was said to have been undertaken "to develop an end to end health and fitness system that comprised mobile phone and PC based applications that individuals could use to easily and cost-effectively find on the internet health and fitness programs that meet and track and help them attain their health and fitness objectives…". Mr Nicolau stated that through R&D activities that his company Akyman Investments Pty Ltd (Akyman) had performed for the applicant, six algorithms were created and tested for the applicant, comprising - (a) a health and fitness algorithm; (b) a health and fitness dietary algorithm; (c) an anti-collision algorithm; (d) a smoothing algorithm; and (e) two calorie measurement algorithms (intake and consumption). 76 Mr Mark Nicolau then described the activities that he said were undertaken by reference to the core activities for the 2014 and the 2015 years. The statement was accompanied by over 860 pages of attachments which included a diagrammatic plan, the index that Mr Werner Nicolau had also produced to the research and development manuals, and tables that referred to different algorithms in their titles. A very large number of conceptual diagrams and flow charts were included in the attachments. The "User Requirements & Feasibility Study" to which I referred above was produced again, and other documents bearing the same title with respect to "Find Phone mobile appellant" and "ZAP Contact mobile APP" were also produced. Included in the attachments were miscellaneous emails and a number of handwritten notes which appeared to record ideas, musings, and mathematical formulae including matrix equations. A copy of the same document titled "Health & Fitness System existing service providers" to which I referred above was produced again, though under a different coversheet which identified it as version "V.1". There were a number of other documents described as industry surveys on various matters. A document titled "Health & Fitness mobile system conceptual design" was produced. A number of documents that were described as specifications or detailed designs of various systems were produced, comprising hundreds of pages. A number of documents appearing to be test plan forms identifying "N Nicolau" as the tester, and the "personal diet" documents to which I referred above were produced again, including the document referring to Chinese food and pizzas that I extracted at [71] and [72] above. A table titled "Fitness Test Records" for Werner Nicolau was produced, recording data such as weight, body mass index, heart rate, and exercise activity for various days during the period 12 May 2014 to 29 September 2017. Some conceptual design, specification and feasibility study documents related to what was described as a "ZAP Contact mobile APP" were also produced. 77 Mr Mark Nicolau prepared a second statement dated 17 May 2018. This statement referred to a meeting on 4 June 2015 between Mr Werner Nicolau, Mr Mark Nicolau, and representatives of "AusIndustry", including Dr Jessie Hiu Kiu to whom I will refer later. Attachments to the second statement included a 45 page report prepared by Mr Mark Nicolau and a 12 page spreadsheet containing information about the claimed activities. The report spoke to the applicant's health and fitness system and to the claimed core R&D activities for the 2014, 2015, and 2016 years, and addressed the respondent's findings. The report contained summaries of information set out in a number of tables spanning 23 pages. In respect of the 2016 year, Mr Nicolau's report stated at page 7 - 2015/2016 • The test results from the previous year enabled us to consider additional ECO system parameters affecting the calorie intake and burning rates and improve our self-learning AI engines. Wind, Grip, Gradient, etc affect the calorie burning rate. At the same time blood pressure, sugar level, cholesterol level and stress level affect the calorie intake rate. • The protection of end-users' confidential health information became the key objective in the 3ʳᵈ year of H&F project. We have to determine what confidential information to protect and how to protect it considering that end-users, medical practitioners, dieticians and fitness instructors are equipped with off-the-shelf non-secure devices such as mobile phones and personal computers. In this respect we have considered the tokenization of access identification and the encryption of confidential information. Since end-users' devices are non-secure we had to determine how to use Cloud security and Tokenization of critical data to protect information access and exchange. • Now that the end-to-end security aspects of the H&F system have been defined we had to consider the monetary implications associated with the provision of mobile and PC based applications to end-users, medical practitioners, dietician and fitness instructors. We have designed a Cloud based payment system to enable the purchase of H&F applications and on­ going maintenance fees. The problem to be resolved was the processing of access requests and payments initiated by non-secure devices. After considering a number of retail payment solutions we had to adopt the Point Over End (POE) solution where both sender and recipient are located in the Cloud and sharing a unique security module for encryption/decryption, authentication/verification and signing/un-signing. 78 A third witness statement of Mr Mark Nicolau dated 30 July 2018 was filed by the applicant. This statement recounted that Mr Nicolau had requested a dietician, Mr Surdut, to provide an expert report and that he had declined. Mr Nicolau stated that he had consulted Mr Surdut on the referral of his general practitioner on four occasions over the course of six months in late 2013 and early 2014. He recounted that he lost weight after following the advice of Mr Surdut, and incorporated that advice into his research and development of the health and fitness system, and applied the "methodology that was developed from my consulting Mr Surdut" to his other "test subjects" whom he identified as his wife and his two sons. Mr Nicolau stated that documents that he prepared in 2014 and 2015 and which he attached were prepared with the benefit of Mr Surdut's advice. Mr Nicolau also attached a letter that he wrote to the Tribunal dated 8 July 2018 requesting that the Tribunal interview Mr Surdut.

The respondent's written evidence 79 The respondent relied on reports of two expert witnesses: Dr Deborah Kerr, and Dr Jean-Guy Schneider. 80 Dr Kerr, in her first report which was dated 7 March 2018, stated that she was a research academic with a particular interest in assessing diet and diet behaviours using technology. She stated that she was employed as Associate Professor, Nutrition, Food Science and Dietetics, at Curtin University, and was an accredited practising dietitian with qualifications in nutrition and dietetics. Dr Kerr referred to her extensive experience in applying dietary assessment methods such as food records and food frequency questionnaires. She stated that she had a special interest and expertise in the use of technology, including the development of mobile dietary assessment applications to assess diet. She stated that she was currently conducting a large intervention study that used a fitness tracker app and a mobile food record app, and that she was a named inventor on two US patents for a mobile food record app covering the methods and the systems. 81 The respondent provided Dr Kerr with a number of documents that were listed in the letter of instructions that was attached to the report. Those documents included but were not limited to the applicant's R&D Tax Incentive applications for the 2014 and 2015 years, the first witness statement of Mr Werner Nicolau with some of the attachments to that statement, and the first witness statement of Mr Mark Nicolau together with some of the attachments to that statement. Dr Kerr was asked to address questions about the applicant's claimed core R&D activities 1.1 and 1.2 for each the 2014 and 2015 years. 82 At the commencement of her first report, Dr Kerr made the following overarching comments, where "UVI" is a reference to the applicant - After undertaking a review of the documentation provided to me, my key points are: • UVI has not carried out scientific experiments that follow the principles of established science in the fields of nutrition, exercise science and mobile applications to assess diet and exercise. • UVI have not provided evidence that they have developed anything new or innovative and instead confirm or re-state what was already known in the field at the time. • Throughout the documentation, UVI have used terminology either incorrectly or not in the correct context (for example, the use of terms "calorie consumption", "algorithms" as explained in paragraph [106]. • Details of scientific experiments are lacking and show no linking between the objectives, methods, the scientific findings and conclusions. In addition, the documentation of the experiments described in the testing records (for example, documents [MN-4L] and [MN-4M] is of poor quality and does not follow the principles of established science). • The evidence provided on the "experiments" conducted by the Applicant appears to be on one or two participants detailing their daily food intake, exercise routine and weight loss over time (for example see [MN-4L], [MN-4M]). The outcomes of these experiments could easily be determined by a competent health professional in the field of nutrition and exercise science who carries out this type of monitoring with clients and interprets the findings to guide them in improving their health and fitness. This is not a scientific method that can address the alleged objectives or provide scientific proof in support of the hypotheses as described. 83 In relation to each of the four claimed core activities that Dr Kerr was asked to address, some of the principal opinions expressed in the report were as follows - (a) Dr Kerr considered Activity 1.1 in respect of the 2014 and 2015 years together, as she could find no evidence that improvement in the fitness monitoring algorithm occurred during the specified time period. Dr Kerr stated that there was essentially no evidence that the applicant had conducted experiments employing scientific method or generating new knowledge, and no evidence that the principles of established science had been followed. Dr Kerr stated that the incorrect use of terminology throughout the documentation showed a lack of understanding of the established body of science underpinning nutrition and exercise metabolism. Further, there was simply nothing novel about the activities that the applicant was undertaking. There was nothing in the material that showed any evaluation of testing by the applicant, and there was no evidence of any conclusions being reached by the applicant following any testing of its hypothesis. (b) As to Activity 1.2 in respect of the 2014 and 2015 years, Dr Kerr considered them together, stating that she could find no evidence of two distinct activities. Dr Kerr stated that the documents that she had reviewed did not contain evidence that the applicant had developed any algorithm to measure dietary intake, or any control algorithms, and that on the evidence available the applicant appeared only to have synthesised existing knowledge and literature in the field. Dr Kerr stated that from her experience as a nutrition scientist who has been involved in the development of mobile apps that she could find no evidence that the development of a calorie intake app had been undertaken. Further, Dr Kerr stated that the applicant had not provided any evidence to indicate that these registered activities were undertaken. As to the records of food intake, Dr Kerr stated that the applicant undertook monitoring of one or two individuals by recording of their diet using a paper-based food record; however neither of these "experiments" was relevant to either of the activities as they were purely observational and anecdotal on a single person and did not provide measurable empirical evidence linked to the relevant hypothesis, and that keeping a food log was nothing new. (c) Overall, Dr Kerr stated that based on the material provided she did not accept that the experimental activities employed a scientific method, and that she did not accept that anything was actually done by the applicant beyond generating documents based on publicly available information, and some limited monitoring of diet and exercise by one or two individuals. 84 Dr Kerr provided a second report dated 6 August 2018 consequent upon a request by the respondent's solicitors to consider further material. The further material was listed in the letter of instructions and included but was not limited to the applicant's statement of facts issues and contentions in reply, the second witness statement of Mr Mark Nicolau and in particular his report that was attached to that statement, the third witness statement of Mr Mark Nicolau that referred to his personal consultations with his dietician Mr Surdut, and a letter dated 4 July 2018 provided by Mr Surdut. The key point made by Dr Kerr in her second report was that after undertaking a review of the additional documents she could find no evidence that changed, altered, or qualified her opinions in her first report. 85 Dr Schneider, in his first report dated 7 March 2018, stated that he had over 26 years' experience in software engineering, and had post-graduate degrees from the Institute of Computer Science and Applied Mathematics at University of Berne in Switzerland. At the time of writing the report Dr Schneider held the position of Associate Professor, Faculty of Science, Engineering and Technology at the Swinburne University of Technology in Melbourne. Dr Schneider was provided with a large number of documents, including but not limited to the applicant's R&D Tax Incentive applications for the 2014 and 2015 years, the first witness statements of Mr Mark Nicolau and Mr Werner Nicolau, and a number of the attachments to those statements. Dr Schneider stated that while a vast amount of material had been provided by the applicant, few documents were actually relevant to the activities being claimed or were helpful in assisting him to understand what the applicant was actually doing. Dr Schneider said that the materials provided contained a significant amount of duplication, with various documents being provided multiple times, and multiple versions of the same document being included. Some documents were wrongly labelled, which in context I take to mean wrongly titled by the applicant. A number of documents in Dr Schneider's opinion appeared to be irrelevant to the claimed activities, including those relating to the Zap Contact App, and there were very few relevant documents in all of the material provided. Dr Schneider stated that there was no evidence in the relevant documents that the claimed activities went beyond the concept stage, and no evidence that any testing took place beyond the applicant's claim that it undertook "extensive testing". In Dr Schneider's opinion there was an array of deficiencies in the information in the documents. 86 Dr Schneider considered activities for the 2014 and 2015 years together. As to the claimed core activities that Dr Schneider was asked to address, some of the principal opinions expressed in his report were - (a) As to claimed activities within Activity 1.1, Dr Schneider stated that the development of the algorithm was beyond his expertise, and he looked at the questions as a software engineer who nonetheless knew what level of detail had to be contained in an algorithm to write executable code. Amongst other things, Dr Schneider said that there was no evidence that the activity moved beyond the conceptual stage. No source code or executable binary code had been provided, and there was no evidence of any prototyping, testing or experimentation in relation to software development. There was no evidence that the relevant algorithm was developed to a level of detail suitable for implementation. As to the test records that were completed by hand and to which I referred earlier, Dr Schneider stated that any conclusions drawn from those records were very unlikely to withstand scrutiny from the scientific community, as no generally accepted scientific method was used to demonstrate the validity of the stated hypotheses. From a software engineering perspective there was no evidence that the applicant was trying to resolve a technical uncertainty, and it seemed unlikely that could have been the case anyway. (b) As to claimed activities within Activity 1.2, Dr Schneider was of the opinion that as with Activity 1.1, nothing moved beyond the conceptual stage. Dr Schneider stated that he was unable to describe the activities, because there was no evidence of what the health monitoring algorithm looked like, or what if any software development took place. (c) As to claimed activities within Activity 1.3 for the 2014 year, being the conceptual design and evaluation of a potential implementation of cloud based decision support systems, Dr Schneider expressed the opinion that there were no specific technical unknowns that had to be tested or investigated, and that from a conceptual perspective there was nothing particularly novel about the proposed system. Dr Schneider stated that by 1 July 2013, decision support systems were established technology and cloud based systems had been around for at least a decade. There was no evidence of any experiments or testing, or in fact that the "cloud based decision support system" was ever created. (d) As to the claimed activities within Activity 1.3 for the 2015 year, being improvements of cloud based anti-collision systems with smoothing algorithms, Dr Schneider stated that the idea of these algorithms was not new. In view of the lack of detail and that the respective flowcharts provided were incorrect, one could not ascertain whether the applicant ever went past the conceptual stage. There was no evidence of specification, design, implementation, or testing of the algorithms. There was nothing novel about smoothing algorithms, which had been studied for decades, and similar search algorithms could have been adapted by the applicant. Dr Schneider pointed to some inconsistencies in the activity logs in relation to the anti-collision engine, and stated that the documents did not explicitly state what specific R&D activities this activity was supposed to include, and that the material provided did not describe any specifics of experiments or testing that was conducted. Other observations made by Dr Schneider pointed to the lack of information or details, and the lack of evidence. 87 Dr Schneider provided a second report dated 6 August 2018. That report was in response to a request by the respondent's solicitors to consider further material. As with the further material given to Dr Kerr, the further material provided to Dr Schneider was listed in the letter of instructions and included but was not limited to the applicant's statement of facts issues and contentions in reply, the second and third witness statements of Mr Mark Nicolau and his report that was attached to the second statement, a letter dated 4 July 2018 provided by the dietician Mr Surdut, and complete versions of a number of documents previously provided. The principal opinions expressed by Dr Schneider in his second report were that that additional materials had not led him to change the opinions expressed in his first report, and that he was unable to identify any further evidence of research and development activities carried out by the applicant during the 2014 and 2015 years.

The hearing before the Tribunal 88 The hearing before the Tribunal took place over three days: 29 and 30 October, and 8 November 2018. The applicant corporation was not represented at the hearing before the Tribunal by a legal practitioner, but was represented by Mr Werner Nicolau, and his father, Mr Mark Nicolau, both of whom also gave evidence before the Tribunal. Mr Mark Nicolau was also the sole director of Akyman to which I referred earlier, which was registered as a research services provider under s 29A of the IR&D Act for the relevant years. It was part of the applicant's claim that Akyman had provided services which contributed to the undertaking of the registered R&D activities. The respondent was represented at the hearing before the Tribunal by counsel, instructed by solicitors. 89 The applicant provided to the Tribunal a written opening dated 29 October 2018. As with the statement of facts issues and contentions in reply, the applicant's written opening to the Tribunal raised a number of claims of error by the respondent as if on judicial review, and sought to put in issue whether the applicant had undertaken R&D activities for the 2014, 2015, and 2016 years. 90 In his oral opening before the Tribunal, Mr Mark Nicolau representing the applicant made some incidental references to the 2016 year. However, in the course of ruling upon an objection to evidence towards the conclusion of the opening, the Tribunal stated that the 2016 year was not before the Tribunal, which was consistent with the rulings it had made at the directions hearings on 14 September and 23 October 2018. The particular ruling on evidence was to exclude the brief second statement of Mr Werner Nicolau dated 7 August 2018 which referred to the applicant's request to the respondent to review the decision in respect of the 2016 year. Counsel for the respondent made a brief opening, referring the Tribunal to the relevant legislation. 91 Over the course of three hearing days all four witnesses were cross examined at length. The two expert witnesses were interposed without objection by the applicant. The Tribunal allowed both Mr Mark Nicolau and Mr Werner Nicolau to cross-examine Dr Kerr. 92 The Tribunal was actively engaged in the course of the hearing, seeking clarification of the evidence from time to time, disallowing certain questions, disallowing the tender of some documents, and giving general guidance to the applicant's representatives about the issues that were before the Tribunal to encourage them to focus on those issues. In particular, in response to attempts by the applicant to identify errors in the respondent's decision-making, the Tribunal stated that the hearing was a de novo hearing which the Tribunal described to the applicant's representatives as involving a fresh decision. The Tribunal conveyed this to the applicant's representatives on a number of occasions. Early in the hearing, the Tribunal stated to Mr Mark Nicolau that it was for him to put the case again. The Tribunal also assisted the applicant by taking Mr Werner Nicolau through his evidence-in-chief. 93 At the conclusion of the hearing on 8 November 2018, the Tribunal directed that written submissions be filed by the parties. The applicant filed submissions with the Tribunal dated 23 November 2018 that were signed by Mr Werner Nicolau on behalf of the applicant comprising 64 numbered paragraphs and 38 numbered endnotes within 17 pages. Within those submissions the applicant contended, amongst many other things, that information relevant to the 2016 year was probative of the nature of the project that was being developed in 2014 and 2015 - 20. In order for the Tribunal to properly understand the Health and Fitness System in its entirety, the Tribunal needs to include consideration of all of the components of the Health and Fitness System including the transaction processing, exchange of information, security, and integration of the functions required to process the Health and Fitness transactions. While the Tribunal proceeding has been limited to the years ended 30 June 2014 and 30 June 2015 (and has excluded the 2016 year), this should not limit consideration of the Health and Fitness System in its entirety. That is, the Tribunal needs to consider all of the components of the Health and Fitness System regardless of when the components were being developed. 21. The proposed product required the bringing together of several existing and new technologies in an innovative way. The Health and Fitness System did not spring fully formed in 2014 or 2015. The information on the Health and Fitness System relating to the year ended 30 June 2016 is probative of the nature of the R&D project into the Health and Fitness System that was being developed in 2014 and 2015. 94 The applicant then submitted to the Tribunal that a number of necessary essential components designed in the 2014/2015 year for use in the Health and Fitness System were created and claimed in the 2016 year, identifying a transaction processing system, applications for exchange of information being the "Find Phone mobile APP", the "Prescription Monitoring mobile APP", and the "ZAP Contact mobile APP", a security system for the exchange of medical information, and the integration of functions required to process the health and fitness transactions, which was described as follows - In addition to the above stated by-products and prototypes, the Cloud based Health & Fitness System required the integration of various Decision Support Systems (DSS) and Artificial Intelligence (Al) algorithms (search engines, non-collision, smoothing, etc), and the testing of each algorithm individually and also in conjunction with the other 5 algorithms. The linking was conceptually developed but had not been coded at that time. A comprehensive Test Plan was developed for the Health and Fitness System, however, due to the lack of funding, the development of test drivers and stubs (the stubs contain the data to be tested and the expected results), and the testing of each function with the context of normal, boundary and transient conditions was shelved. (Footnote omitted.) 95 The respondent filed written submissions of counsel in response dated 6 December 2018 comprising 67 numbered paragraphs with 95 numbered footnotes within 15 pages. 96 The applicant replied to the submissions of the respondent by written submissions dated 18 December 2018, which were again signed by Mr Werner Nicolau on behalf of the applicant. The reply submissions were 12 pages in length, including an attached chart, and contained 110 endnotes referring to documents filed with the Tribunal and to transcript references.

The Tribunal's statement of reasons 97 The Tribunal affirmed the decision under review. The Tribunal's statement of reasons comprised 33 pages, containing 100 numbered paragraphs. Within those paragraphs, the Tribunal copied without attribution almost the entirety of the respondent's written outline of submissions dated 6 December 2018, and included them as part of the Tribunal's own reasons. However, the paragraphs of the respondent's submissions that were copied were not presented in the same order in the Tribunal's reasons, and they were placed amongst other paragraphs that were not copied. 98 Prior to the hearing of the appeal, I made an interlocutory order that the respondent file a copy of the Tribunal's reasons that highlighted in yellow those passages of the reasons that the respondent accepted had been copied verbatim, or at least substantially, from the respondent's submissions to the Tribunal, and that the respondent also file a copy of its closing submissions to the Tribunal which had corresponding highlighting. Those documents were helpful. Most of the respondent's submissions to the Tribunal, from start to finish, were copied verbatim. However, as I have mentioned, the paragraphs were not all copied in the same order as they appeared in the respondent's submissions to the Tribunal. A small number of paragraphs, while not copied verbatim in their entirety, were copied for the most part. The substance of the two concluding paragraphs was replicated in the Tribunal's reasons. Of the 67 numbered paragraphs of the respondent's written submissions, 65 were copied by the Tribunal in this way. 99 Viewed from the perspective of the Tribunal's statement of reasons, of the 100 paragraphs, about 65 paragraphs and their footnotes comprise unattributed copying of the respondent's written submissions. Those paragraphs that were not copied from the respondent's written submissions included - (a) [3] to [20] setting out uncontroversial matters such as the identification of the applicant's registered activities, and the relevant statutory provisions; (b) a framing of the issues at [47]; (c) identification of the witnesses who gave evidence at [50]; (d) summaries of the applicant's submissions at [52] to [59], including observations by the Tribunal at [52], [53], [57] and [59] in response to the submissions that were put; (e) references to and summary of the applicant's submissions in reply at [60] and [61]; (f) identification of the Tribunal's conclusion at [62]; and (g) a comment about the oral testimony of the applicant's witnesses at [70]. 100 Otherwise, the Tribunal's consideration of the evidence comprising [63] to [69], and [71] to [97] were copied essentially verbatim from the respondent's submissions, although as I have stated there was some re-ordering of things. The copying included the footnotes in the respondent's submissions, most of which were reproduced as footnotes in the Tribunal's statement of reasons. The footnotes included detailed references to authorities, and to the documentary evidence and transcript. The copying extended to the reproduction of the following typographical errors that were contained in the respondent's written submissions - (a) at [37], which was copied from [22] of the respondent's submissions, "field" was misspelled as "filed" in a quotation from [2.13] of the explanatory memorandum to the Tax Laws Amendment (Research and Development) Bill 2010 (Cth); and (b) at [71], which was copied from [39] of the respondent's submissions, "ZAP contact mobile app", which was referred to on a number of occasions in the evidence before the Tribunal, was misspelled as "ZAP contract mobile app" (underlining added). 101 However, balanced against the above are other typographical errors in the respondent's submissions that the Tribunal did not adopt - (a) footnotes (62), (76), and (89) of the Tribunal's reasons contain corrected references to documents before the Tribunal that were cited in footnotes (51), (65), and (78) of the respondent's submissions; and (b) at [81], the Tribunal corrected the date on which the report of an expert, Dr Kerr, had been filed with the Tribunal. 102 Because of the root and branch nature of the applicant's challenge to the Tribunal's reasons, it is necessary to refer to its reasons at some length. Putting the question of copying to one side, the Tribunal's reasons were as follows. 103 At [2], the Tribunal identified the reviewable decision of the respondent which affirmed that none of the applicant's activities registered under s 27J of the IR&D Act for the two years the subject of the review satisfied the definition of "R&D activities" in s 355-20 of the ITAA 1997. 104 At [4], the Tribunal set out the activities for the 2014 and 2015 years, to which I referred at [23] and [30] above. 105 At [5] to [20], the Tribunal set out an overview of the relevant provisions of the ITAA 1997 and IR&D Act, substantially adopting the overview given by Forgie DP in Moreton Resources Ltd and Innovation and Science Australia (Taxation) [2018] AATA 3378 at [4]-[19], which the Tribunal cited. An appeal from the Tribunal's decision in Moreton Resources Ltd was subsequently allowed, but nothing turns on that for present purposes: Moreton Resources Ltd v Innovation and Science Australia, cited above. 106 At [21]-[29], the Tribunal made further references to the statutory framework. 107 At [36]-[41], the Tribunal addressed the first limb of the statutory test in s 355-25(1)(a) of the ITAA 1997 in the following terms - 36. To be an eligible core R&D activity, the outcome of the activity cannot be known or determined in advance without conducting a systematic progression of experimental work based on principles of established science. 37. The words 'cannot be' in s 355-25(1)(a) of the ITAA 1997 are absolute. For an experimental activity to qualify as a 'core R&D activity', there must be a high level of uncertainty as to whether the outcome can be known or determined in advance based on current knowledge, information or experience, and prior to any experimentation taking place. Further, s 355-25(1)(a) requires that the 'only' way in which the outcome of the activities can be determined is through the application of, in effect, the scientific method (see paragraph 36 above). The 2010 R&D EM explains at paragraphs 2.13 to 2.15 that this requirement is a 'threshold… knowledge gap' and the 'threshold will not be met if the knowledge of whether something is scientifically or technologically possible, or how to achieve it in practice, is deducible by a competent professional in the filed [sic] on the basis of current knowledge, information or experience'. Further, there must be 'a clear risk that the outcome of the experiment will not be the desired one'. 38. Both the way in which s 355-25(1)(a) of the ITAA 1997 is drafted, and the reference to the 'competent professional in the field' in paragraph 2.13 of the 2010 R&D EM, makes it clear that the uncertainty of outcome associated with the activity must be objectively determined. It is not enough that unqualified individuals without any knowledge, information or experience of the principles of established science cannot determine the outcome of the activity before undertaking it. Similarly, it is not enough that the product sought to be made is 'new' if it could be produced by a competent professional. The question is whether the outcome of the activity could or could not be known or determined in advance based on current knowledge information and experience, and the only way in which the outcome can be achieved is through a systematic progression of work that conforms with subparagraphs 355-25(1)(a)(i) and (ii) of the ITAA 1997. 39. The juxtaposition of the words 'cannot be' and 'but can only be determined' in s 355-25(1)(a) of the ITAA 1997, when considered together with the objects of Div 355 and the extrinsic materials, support the logical interpretation of s 355-25 that requires that there is, objectively, a high degree of uncertainty about the outcome of the activities, and that the risk must be scientific or technical in nature. Paragraph 2.18 of the 2010 R&D EM makes this clear when it states: "The need to employ the scientific method also reflects the degree of novelty in the ideas being tested. That is, the knowledge being sought must go beyond validating a simple progression from what is already known and beyond merely implementing existing knowledge in a different context or location…." 40. For an activity to be a core R&D activity, the Applicant must establish that it developed a hypothesis capable of being tested by conducting an experiment. The hypothesis must be focused on one or more technical or scientific areas of uncertainty that can only be resolved through experimentation. A hypothesis focused on commercial objectives is not sufficient. 41. The Applicant must also establish that it observed and evaluated the results of its experiment. This requires, at a minimum, that the Applicant observed, measured and recorded the information and results relating to the experiment, before analysing and evaluating those results. The Applicant must therefore demonstrate that the systematic progression of work led to logical conclusions as to whether the hypothesis is right or wrong. (Footnotes omitted.) 108 At [42]-[44], the Tribunal addressed the requirement in s 355-25(1)(b) of the ITAA 1997 that an activity be conducted for the purpose of generating new knowledge - 42. Section 355-25(1)(b) of the ITAA 1997 further requires that an activity be conducted for the purpose of generating new knowledge (including knowledge in the form of new or improved materials, product, devices, processes or services). The reference to 'the purpose' in s 355-25(1)(b) is not a reference to a dominant or prevailing purpose. However, the use of the definite article 'the' indicates that the relevant purpose must be substantial. 43. The purpose for which activities are carried on 'is the object or aim for which they were carried on at the time' and is not determined 'according to a rationale developed at a later time to explain why those activities were carried on'. Moreover, the purpose of the claimed activity must be proved by evidence (which will almost always be in the form of contemporaneous documentation). 44. Under the express terms of s 355-25(1)(b), the knowledge being sought must also be 'new'. 'New knowledge' in this context means knowledge not already available in the public domain at the time the activities are conducted, in the relevant technology, on a reasonably accessible worldwide basis. It does not mean knowledge that is new only to the Applicant. (Footnotes omitted.) 109 At [47], the Tribunal framed the issues before it as follows (formatting as per the Tribunal's reasons) - 47. The Tribunal is to determine these three issues: (1) Did the Applicant conduct the Activities? Has the Applicant proved, on the balance of probabilities, that it conducted the activities as registered by it in each of the years ending 30 June 2014 and 30 June 2015? (2) If the Applicant did conduct one or more Activities that it registered as "core R&D activities", do any one or more of those Activities satisfy the requirements for "core R&D activities" in s 355-25 of the ITAA 1997? (3) If the Applicant conducted any "core R&D activities", and proved that it also conducted other Activities, do any of those other Activities satisfy the requirements for "supporting R&D activities" in s 355-30 ITAA 1997? If the answer to question 2 is "yes" in respect of any one or more Activities registered as "core R&D activities", and there are other Activities that the Applicant proved on the balance of probabilities were conducted in the years ending 30 June 2014 and 30 June 2015, then do any of those other Activities meet the requirements for "supporting R&D activities" under s 355-30 ITAA 1997? The Applicant's claim fails for any Activities in respect of which the answer to the above questions is "no". 110 At [48], the Tribunal elaborated on the issues that it framed, stating (inter alia) - 48. Hence, the fundamental issue for determination by the Tribunal is not whether the Applicant had an idea that, when described in general terms, could be characterised as being 'innovative'. The question is whether the actual activities were carried out by or on behalf of the Applicant as registered in the Relevant Years and, if so, whether any one or more of those registered activities was an 'R&D activity', as defined. … 111 At [51], the Tribunal made a finding about the activities that the applicant had registered - 51. The way in which the Applicant registered its activities for the Relevant Years is set out above. Critically, the detail of the registrations makes it clear that the experimental activities associated with registered activities 1.1 and 1.2 in the Relevant Years were exercise-related for the purposes of determining calorie consumption for use in the proposed 'fitness algorithm' and diet-related for the purposes of determining calorie intake for use in the proposed 'health algorithm', respectively. The Applicant cannot now endeavour to disavow the description of the activities as registered, by attempting to recharacterise its activities as purely software engineering related, as discussed below. The Tribunal's review is constrained by the focus in s 27J of the IR&D Act on the registered activities. (Footnotes omitted.) 112 At [52]-[59], the Tribunal summarised the submissions of the applicant - 52. The Applicant submitted its system would integrate the user's health and fitness objectives, with their physical characteristics and their medical history. It would communicate with the gym equipment, smartwatches and the bathroom scale. It would communicate with a secure, contemporary, database of scientific knowledge. It would seek an optimal solution for the user. It would track results and work pragmatically with user preferences, to take the path that takes the individual to his or her goal. As discussed above, far more than an idea or concept is required to meet the criteria in s 355-25 ITAA 1997. 53. The Applicant submitted that it had an idea to build something new in 2013. In its written closing, it submits that the 'key to this case' is catching what the Respondent has missed. However, clearly it was for the Applicant to put before the Tribunal the necessary material required to support a finding by the Tribunal in its favour. 54. The Applicant submitted that the research before the Tribunal produced several algorithms which were explained in the Functional Specifications and Detailed Design documents. It submitted that the experimentation completed was based on principles established in science, being software engineering, and not fitness or dietary domains of the health industry. 55. The Applicant submitted in software engineering the experiments are carried out for each development phase starting with the conceptual design, functional specifications and detailed design; where concepts are tested for viability, functionality and logic and followed by test planning, source code testing and alpha and beta testing. It was claimed the R&D Manuals contain the evidence of the work progress in accordance with principles of established science. 56. The Applicant submitted that the independent experts, called by the Respondent, were confused about the project scope in stating that the Applicant was trying to develop another fitness or dietary program and not the world's first integration of existing programs with end-user's personal profile, health status and objectives. 57. The Applicant was critical of Dr Kerr's expertise as a dietician, rather than in software engineering. She was said to be 'unable to see the evidence of the software engineering work that was in front of her'. However, she did testify, relevantly, "based on the evidence that I had, there is a - you have a concept and an idea that you've presented, however, there is no experimentation or evidence to support the work actually exists". 58. The Applicant submitted that Dr Kerr and Associate Professor Schneider didn't know what its system was about. "It was an innovative system, such as done by Apple and Google. An iPod is simple, once you've seen it. Not before. The Applicant's Health and Fitness System had not been seen before. It is all easy after it has been done. Then you don't need research". 59. By reason of s 27J of the IR&D Act, it is necessary for the Tribunal to focus upon the registered activities as opposed to the generality of ideas and determine what if any relevant research and development work, in compliance with the IR&D Act, as analysed above, has been completed by the Applicant during the Relevant Years. 113 At [60], the Tribunal expressly referred to the applicant's closing submissions in reply - 60. The Applicant, in its written Closing Submission in Reply, maintained, again, that it is the Respondent which has "failed to substantiate its case". The Applicant purported to assert discrepancies in the evidence of the Respondent's experts in respect of their evidence on design and specifications, innovation and experimental activities. 114 The Tribunal concluded its summary of the applicant's submissions, by extracting and summarising a part of the applicant's closing submissions in reply - 61. The Applicant submitted that "the systematic progression of work that proceeds from hypothesis to experiment, observation and evaluation, and leads to logical conclusions is applied in the development of its Health and Fitness System as a whole", and the registered activities which form part of that system are evidenced by: (a) algorithms; (b) the five Tables at Tab 9.3 on page 492 of the Tribunal Book (Ex R4 pp 492-496); (c) the documents referenced in the five Tables; (d) the existence of different versions of such documents; and (e) that part of the specification and design of transaction processing and the exchange of information, components of the Health and Fitness System which was commenced in the 2015 year falling within the Relevant Years. 115 The Tribunal then made an overarching factual finding at [62], expressed in the following terms - 62. However, on balance, after a thorough analysis of the documentary evidence in this matter together with consideration of the oral testimony, the Tribunal finds that insufficient evidence exists to indicate that any of the registered core R&D activities were conducted as discussed below. 116 In elaboration of its overarching conclusion, the Tribunal referred to the witness statements that had been filed and admitted into evidence, and stated - 66. These witness statements contained insufficient direct evidence of what was actually done by the Applicant, or by Akyman, in the way of experimental activities in the Relevant Years at a level of detail to support the finding by the Tribunal that the registered activities were either 'core R&D activities' or 'supporting R&D activities', as defined. Rather, the witness statements comprise conclusory assertions expressed in general terms. 67. The witness statements do attach documents, but many of those documents are irrelevant, as discussed below. Of the remaining documents, which do appear, on their face, to relate to the Applicant's proposed 'health and fitness program', there are multiple versions of these documents, most of which show very little progression in terms of the work purportedly done by the Applicant - and none of which record anything in the way of actual 'experimental activity' in the form of a systematic progression of work based on the principles of established science, as required. 68 Further, such documents were demonstrated to be unreliable. According to the Applicant's witnesses, the dates on the documents did not necessarily correspond to the actual date of the document. There were also numerous errors in the documents. For instance, by Mr Mark Nicolau's own admission, the invoices produced by Akyman included work not actually performed. 69. The Applicant's witnesses also gave answers during cross examination that were often contradictory, confused or evasive, and not based on any independent recollection of the purported work actually performed on the Applicant's registered R&D activities. By way of specific example: (a) Whereas Mr Werner Nicolau gave evidence that his mother was a test subject, Mr Mark Nicolau characterised his wife's (being Werner's mother) involvement as being in an administrative capacity only. (b) Mr Mark Nicolau could not confirm when he first met with the dietician, Mr Surdut, nor whether he met him four or five times. (c) The evidence concerning when work began on the health and fitness 'field tests' (or 'pilot program') was inconsistent and confusing, with Mr Mark Nicolau ultimately admitting he had no independent recollection of when work began. Similarly, neither witness appeared to know when testing ended. 70. The oral testimony of the Applicant's witnesses did not add any of the requisite detail to the unsatisfactory state of the documentary evidence relied upon by it. (Footnotes omitted.) 117 At [71] to [75], the Tribunal referred to material before it that it considered to be irrelevant to its determination - 71. Much of the material relied on by the Applicant before the Tribunal was irrelevant to the Tribunal's review. This included all the material relating to its registered activities in the 2016 year, in addition to documents relating to the Near Field Communication (NFC) work performed by Akyman on behalf of a different company, as well as the transaction card processing system, 'Find Phone Mobile App', the 'ZAP contract mobile APP' and a 'power socket rotative adaptor'. 72. None of these activities are mentioned in the Applicant's registrations for the Relevant Years. They are therefore irrelevant to the question of whether any of the Applicant's registered activities in the Relevant Years were actually conducted either as core R&D activities within the meaning of s 355-25 of the ITAA 1997 or as supporting R&D activities within the meaning of s 355-30 of the ITAA 1997, in either of both of the Relevant Years. The Applicant's continued attempt to rely on irrelevant material, in an impermissible attempt to extend the scope of its activities, only highlighted the lack of material before the Tribunal relating to the actual registered activities for the Relevant Years. 73. Records of personal diets and personal gym programs of the Applicant's witnesses relied on by the Applicant are also irrelevant. 74. In addition, the Applicant persisted in its attempts to rely on notes and documents prepared by staff of the Department during earlier reviews of the Applicant's activities. Notwithstanding that this material was irrelevant to the Tribunal's review of the decision made by the Respondent under s 30D of the IR&D Act. 75. In any event, the decision under review is in no way inconsistent with the conclusion reached by Dr Jessie Hiu Kiu on the original compliance review completed on or around 9 July 2015. By letter dated 9 July 2015, Dr Jessie Hiu Kiu informed the Applicant that its registration had been rated as having a 'medium risk of non-compliance', and that the Applicant had been 'flagged for further review upon the next registration submission'. (Footnotes omitted.) 118 The Tribunal then considered in turn the registered activities for the two relevant years, and made findings of fact at [76]-[97] which I will now summarise.

Activity 1.1 119 At [78]-[82], the Tribunal found that the evidence did not disclose that any experimental activity of the type described in the registrations was carried out by or on behalf of the applicant. The Tribunal found that no more than two or three members of the Nicolau family completed test record forms relating to their exercise. No women participated. The Tribunal held that it was difficult to conclude that a scientific method was followed, that no objective baseline measurements were established, and that the test record forms all recorded highly subjective handwritten annotations. There was no evidence of any meaningful analysis or evaluation, none of the activity was carried out in accordance with the principles of established exercise science, the outcome of registered activity 1.1 could have been known in the relevant years, and that the purported "algorithms" that were produced were so deficient, that one of the experts, Dr Kerr, did not recognise the diagrams as algorithms.

Activity 1.2 120 At [83]-[88], the Tribunal held that the evidence did not disclose that any experimental activity of the type described by the registrations was carried out. The Tribunal stated that, rather, the documents suggested that the "experiment" consisted of no more than Mr Mark Nicolau following a diet that he had been given by a dietician because of his own personal health issues. The Tribunal held that there was no "experiment" as such, because nothing was being tested, and that no scientific method was followed. The Tribunal held that none of the activity was carried out in accordance with the principles of established dietary science, and that the outcome of the activity could have been known in advance in the relevant years. The Tribunal further held that the purported "algorithms" as a result of this activity were deficient, and that there was no objective evidence that the applicant had a purpose of generating new knowledge in relation to the activity.

Activity 1.3 121 At [92]-[96], the Tribunal held that it could not be satisfied that the registered activity occurred in the relevant years at all, and that there was no contemporaneous evidence of what the purportedly experimental activity associated with registered Activity 1.3 was in the relevant years. In particular, the Tribunal held that the compelling aspect of the evidence of one of the experts, Dr Schneider, was that there was insufficient evidence of what was done (if anything) in the way of experimental activity associated with registered Activity 1.3 to evaluate whether the outcome could have been known in advance for the purposes of s 355-25(1)(a) of the ITAA 1997. The Tribunal expressed agreement with the respondent's case that there was no evidence that anything was created, let alone implemented, as a result of activities that were conducted in accordance with the applicant's registrations for the registered activities. The Tribunal further held that there was no evidence that the outcome of the applicant's registered activities could not be known or determined in advance on the basis of current knowledge, information or experience for the purposes of s 355-25(1)(a) of the ITAA 1997. The Tribunal concluded at [97] that, on balance, there was no objective, contemporaneous evidence before it to support any finding or conclusion that the applicant had a purpose of generating new knowledge associated with registered Activity 1.3.

The appeal to this court 122 The applicant relied on an amended notice of appeal filed 15 December 2020 which raised six questions of law, and 18 individual sub-grounds of appeal which were contained within ground 2. The grounds of appeal are set out in the schedule to these reasons. The 18 sub-grounds of appeal within ground 2 overlap, and were not expressed in a way that corresponded to any of the six questions of law that was raised, although by a process of analysis the grounds can be aligned with one or more questions of law. Many of the applicant's grounds and submissions appear to complain of factual findings made by the Tribunal on the basis that the Tribunal failed to take into account "relevant considerations", being other factual matters relied on by the applicant to impugn the Tribunal's decision. There is also a deal of overlap between different parts of the applicant's submissions. I will seek to synthesise the claims made in the grounds of appeal by addressing the six questions of law that were raised, which are the subject-matter of the appeal, and in doing this I have considered all the grounds. Even with the questions of law there is overlap in relation to the subject-matter of the applicant's claims, and this is particularly so with Questions 1 and 5. I will not set out in terms at this stage all the questions of law raised. An indicative summary of the matters that were raised is as follows, adopting the applicant's numbering of the questions of law - (1) Did the Tribunal err by refusing to admit evidence of the applicant's R&D activities for the 2016 year? (2) Did the Tribunal err by failing to take account of the totality of the product being developed by the applicant? (3) Did the Tribunal apply the wrong test to the applicant's R&D activities by incorrectly construing s 355-25(1) of the ITAA 1997? (4) Did the Tribunal err by taking irrelevant considerations into account, or by failing to take relevant considerations into account? This question and the corresponding ground were directed to seven sub-grounds. (5) Did the Tribunal fail to accord the applicant procedural fairness in five respects that the applicant alleged? (6) Was there error of law as a result of the substantial reproduction, without attribution, of the applicant's submissions within the Tribunal's written statement of reasons? 123 The genesis of a number of the matters raised by the amended notice of appeal is a claim that the Tribunal did not consider facts and matters relating to the 2016 year which the applicant contended were relevant to an understanding of what activities had taken place in the anterior years, 2014 and 2015. As I outlined earlier, the applicant filed with the Tribunal an application to review the respondent's decision in respect of the 2016 year out of time, and did not obtain an extension of time until 3 January 2019. 124 The principal relief sought by the applicant was that the matter be remitted to the Tribunal to be heard and decided again. A claim for an order by the court giving the applicant the relief that it sought from the Tribunal was abandoned at the hearing of the appeal.

Question 6 - substantial reproduction by the Tribunal of the respondent's submissions 125 I will consider Question 6 first, because logically it comes first. If by reason of the unattributed copying of the respondent's submissions the Tribunal failed lawfully to discharge its review function, then the review before the Tribunal will have wholly miscarried, and the matter will have to be remitted for re-hearing. I will therefore consider Question 6 before other issues are decided: see in the context of apprehended bias, Concrete Pty Ltd v Parramatta Design and Developments Pty Ltd [2006] HCA 55; 229 CLR 577 at [116]-[117] (Kirby and Crennan JJ) and [172] (Callinan J). 126 Question 6 raises the following questions of law and corresponding grounds of appeal - [Questions of law] … 6. In the circumstances that the Tribunal's reasons for decision substantially reproduced the Respondent's written submissions, without attribution, did the Tribunal make an error of law: a. amounting to a constructive failure to exercise its jurisdiction; b. by failing to exercise independent judgment; c. by acting unreasonably; d. by failing to take into account relevant considerations; e. by taking into account irrelevant considerations; or f. by failing to carry out its statutory functions in a way that would promote public trust and confidence in the decision-making of the Tribunal? … [Corresponding grounds of appeal] 2. The Tribunal incorrectly affirmed the Respondent's decision, as a consequence of the following conduct that amounts to errors of law. … r. In substantially reproducing the Respondents' written submissions, without attribution, as the Tribunal's reasons for decision, the Tribunal made errors of law in: (i) failing to exercise its jurisdiction: (ii) failing to exercise independent judgment: (iii) acting unreasonably: (iv) failing to take into account relevant considerations: (v) taking into account irrelevant considerations: and (vi) failing to carry out its statutory functions in a way that would promote public trust and confidence in the decision-making of the Tribunal. 127 The applicant referred to the extent of unattributed copying by the Tribunal of the respondent's written submissions, which I outlined earlier. The applicant submitted that what the Tribunal said of the applicant was substantially what the respondent had submitted, rather than reflecting any submissions of the applicant. The applicant submitted that the decision "therefore could not give a reasonable observer confidence that the Tribunal fairly dealt with, or considered the applicant's case; or examined all the evidence and made appropriate findings". 128 At the hearing, it was submitted on behalf of the applicant that the wholesale adoption of the primary decision-maker's submissions to the Tribunal detracted substantially from the idea that the Tribunal would act independently, and that the Tribunal failed to conduct its review in a way that promoted public trust and confidence in the decision-making of the Tribunal, as required by s 2A of the AAT Act. It was also submitted that the Tribunal had failed to discharge its statutory function in a way that involved a denial of procedural fairness. 129 In written reply submissions the applicant submitted as follows - A reasonable person could not look at this unattributed copying of the Respondent's Submissions, combined with the fact that the Tribunal made no comparative evaluation of the Applicant's counterarguments in the Applicant's Closing Submission in Reply, and conclude the Applicant had been treated justly or fairly. The Decision was, in effect, the Respondent's Submissions, errors and all. It cannot seriously be asserted that this unattributed copying would not cause a well-founded doubt the Tribunal had independently, properly and fairly executed its statutory task. A fair reading of the Tribunal Decision does not establish, as the Respondent argues, that the Tribunal brought its own independent mind to bear. A fair reading establishes that the Tribunal, in a complex matter requiring rare technical expertise to assess or describe, relied on the Respondent to have correctly assessed and stated every fact, issue and conclusion in a way that could be safely duplicated by the Tribunal. 130 The applicant pointed to aspects of its case before the Tribunal that it submitted required consideration, and submitted that the Tribunal's decision "did not provide refutation of the applicant's arguments". The applicant submitted that the Tribunal's reference to the applicant's submissions were "inaccurate and dismissive" and gave cause to conclude that the Tribunal had not actively applied its own reasoning to the matter, and had not given proper consideration to the matter or dealt with the applicant fairly, had not accorded the applicant procedural fairness, and had not properly performed its fact-finding task. The applicant also complained that the Tribunal's findings as to the presentation and reliability of witnesses called on behalf of the applicant were the product of the adoption of the respondent's submissions. As to specifics, the main matters that the applicant submitted the Tribunal had not considered as a result of copying the respondent's submissions were - (a) the 2016 R&D material; (b) the applicant's written closing submissions in reply; (c) the purpose and objective of the applicant's R&D activities; (d) the contemporaneous evidence of the applicant's R&D activities; (e) circumstances that were favourable to the reliability of the evidence of the applicant's witnesses; (f) evidence of a Dr Jessie Hiu Kiu that was favourable to the applicant; (g) evidence that the applicant had carried out testing; and (h) claimed limitations on the expertise of the respondent's witness, Dr Kerr. 131 The applicant submitted that various other specific errors which were the subject of other grounds of appeal were the result of the Tribunal acting on the respondent's submissions which it had copied into its reasons, which I will identify and consider below. 132 The respondent accepted that of the 100 paragraphs of the Tribunal's reasons, between 65 and 67 paragraphs were copied, more or less, from the submissions of the respondent. The respondent submitted that the fact that the Tribunal adopted a significant portion of the respondent's written outline of submissions did not demonstrate legal error. The respondent accepted that unattributed copying, whether by Tribunal members, academics, or judges is undesirable, but submitted that the copying in this case did not mean that the lawful exercise of jurisdiction by the Tribunal had failed. It was submitted that although there were substantial similarities between the Tribunal's reasons and the respondent's submissions, a fair reading of the Tribunal's reasons established that the Tribunal brought its own independent mind to bear upon the correct or preferable decision on review. The respondent advanced five principal arguments. First, the respondent submitted that the applicant had not identified any critical material before the Tribunal that the Tribunal had failed to take into account. Secondly, it was submitted that the Tribunal had regard to the applicant's submissions as evidenced by [52]-[61] of its reasons, which had not been copied from the respondent's submissions. Thirdly, it was submitted that the Tribunal's conclusions at [62] and [70] of its reasons about the unsatisfactory state of the applicant's documentary evidence had not been copied from the respondent's submissions. Fourthly, it was submitted in relation to the expert evidence that although the Tribunal adopted the respondent's submissions, the findings were congruent with that evidence. Fifthly, it was submitted that the Tribunal's findings about the presentation and reliability of the oral evidence called on behalf of the applicant was also supported by the evidence with the result that there was no error in adopting the respondent's submissions on the issue. 133 In my view, it was not ideal for the Tribunal to express its reasons for decision by reproducing mostly verbatim the respondent's submissions to it, without acknowledging expressly that this was the course that it took. Although it might be apparent what was done, for the Tribunal to produce reasons in this way without attribution is liable to fuel perceptions on the part of applicants for review and others that the Tribunal has not given independent consideration to the matters under review, and does little to meet the exhortation in s 2A(d) of the AAT Act that the Tribunal is to pursue a mechanism of review that promotes public trust and confidence in its decision-making. However, these observations do not address whether there were any legal consequences of the course that the Tribunal took, to which I will now turn. 134 Question 6 raises issues that invite consideration of the following questions - (1) Do the Tribunal's reasons as a whole, by reason of the unattributed copying of the respondent's submissions, give rise to an inference that the Tribunal failed to undertake the review independently and fairly? (2) Related to the second question, do the Tribunal's reasons, by reason of the unattributed copying of the respondent's submissions, give rise to an inference that any particular matter that the Tribunal was required to consider was overlooked? (3) Is there a separate requirement of the lawful exercise of the Tribunal's jurisdiction that it must appear that the Tribunal independently and fairly dealt with the applicant's case such as to promote public trust and confidence in the Tribunal's decision-making? Another way of expressing this question, is to ask whether there is a requirement that administrative justice must appear to have been done.

(1) Do the Tribunal's reasons as a whole give rise to an inference that the Tribunal failed to undertake an independent review? 135 In August 1971, the Commonwealth Administrative Review Committee (the Kerr Committee) recommended the establishment of a general Administrative Review Tribunal that would review the merits of certain administrative decisions made under Commonwealth law. The recommendation was made against the background of the absence of a means of independent merits review of administrative decisions affecting the rights of citizens. Although the Kerr Committee had suggested that the Tribunal might include officers of departments, that suggestion was not accepted by the Committee on Administrative Discretions (the Bland Committee) which reported in October 1973, and the absence of any overt attempt to have representatives of government departments as members of the Tribunal was remarked upon by Senator Everett in debates relating to the Administrative Appeals Tribunal Bill in the Senate: Commonwealth of Australia, Parliamentary Debates, Senate, 3 June 1975 at 2156. 136 Since its commencement in 1976, the jurisdiction conferred on the Administrative Appeals Tribunal has been enlarged. Relevantly, review of taxation decisions was a function conferred on the Tribunal by the Taxation Boards of Review (Transfer of Jurisdiction) Act 1986 (Cth). 137 The AAT Act has a number of features which underscore the independence of the Tribunal from government departments and agencies. Paragraph 2A(d) of the AAT Act contains an exhortation that - In carrying out its functions the Tribunal must pursue the objective of providing a mechanism of review that: … (d) promotes public trust and confidence in the decision-making of the Tribunal. 138 The President of the Tribunal must be a judge of the Federal Court of Australia, and the members of the Tribunal are appointed by the Governor-General: ss 6, 7. Members hold office for the term of their appointment, which is not to exceed seven years: s 8(3). The appointment of members may be terminated only in specified circumstances: s 13. A full-time member must not engage in paid employment outside the duties of office without the President's approval, and a part-time member must not engage in any paid employment that, in the President's opinion, conflicts or may conflict with the proper performance of the member's duties: s 11. 139 Of particular importance is that s 30 of the AAT Act provides that there are to be parties to a proceeding before the Tribunal. Principally, the parties are the applicant for review, and the person who made the decision. The parties are entitled to appear in person, or be represented by another person: s 32(1). There is a requirement that the decision-maker use best endeavours to assist the Tribunal to make its decision in relation to the proceeding: s 33(1AA). There is also a requirement that a party to a proceeding before the Tribunal, and any person representing such a party, must use best endeavours to assist the Tribunal to fulfil the objective in s 2A: s 33(1AB). Subject to a contrary order, the hearing of a proceeding before the Tribunal must be in public: s 35. 140 There is specific content to the Tribunal's obligation to accord procedural fairness. Generally, the Tribunal must ensure that every party to a proceeding before the Tribunal is given a reasonable opportunity to present his or her case and, in particular, to inspect any documents to which the Tribunal proposes to have regard in reaching a decision in the proceeding and to make submissions in relation to those documents: s 39. 141 The Tribunal may exercise all the powers and discretions conferred on the decision-maker, and must make a decision in writing affirming, varying, or setting aside the decision under review: s 43(1). The Tribunal's task is to make the correct or preferable decision on the material before it: Drake v Minister for Immigration and Ethnic Affairs [1979] FCA 39; 24 ALR 577 at 589 (Bowen CJ and Deane J). In other words, the Tribunal's role is to "do over again" what was done by the original decision-maker: Mobil Oil Australia Pty Ltd v Federal Commissioner of Taxation [1963] HCA 41; 113 CLR 475 at 502 (Kitto J), cited in Shi v Migration Agents Registration Authority [2008] HCA 31; 235 CLR 286 at [100] (Hayne and Heydon JJ). 142 The Tribunal is under an obligation to give reasons either orally or in writing for its decision: s 43(2). Where the Tribunal gives the reasons for its decision in writing, the Act requires that those reasons must include its findings on material questions of fact and a reference to the evidence or other material on which those findings were based. This obligation is against the background that there is no obligation upon an administrative decision-maker at common law to give reasons for a decision: Public Service Board of New South Wales v Osmond [1986] HCA 7; 159 CLR 656. 143 It follows from the above features that in conducting a review the Tribunal is distinctly independent of the decision-maker whose decision is subject to review, and who is a party to the proceeding before the Tribunal. The Tribunal is to make its own decision, and is accountable for that decision by the generally open nature of the hearing process, and by the requirement that the Tribunal give reasons. In Minister for Immigration and Multicultural Affairs v Wang [2003] HCA 11; 215 CLR 518, Gummow and Hayne JJ stated at [71] that "[t]he findings of fact that the Tribunal makes are those that it, rather than the claimant, let alone adversarial parties, considers to be necessary for it to make its decision". And as the Full Court (Tracey, Murphy and Mortimer JJ) stated in MZZZW v Minister for Immigration and Border Protection [2015] FCAFC 133; 234 FCR 154 (MZZZW) at [59] in relation to review by the Tribunal under Part 7 of the Migration Act, or any merits review scheme having similar features - … the member constituted to the review will bring her own mind to bear on the issues arising in the review, freed not only from infections such as prejudgment or other bias but from the inevitable constraints on thought, consideration and reflection which flow from the adoption of not only the conclusions of others, but the way those conclusions have been formulated and framed in language. 144 Cases involving the wholesale acceptance by a court or a tribunal of the case of one party over another have been argued and considered through different lenses. In some cases involving judicial decisions, the summary acceptance by a court of the whole of the evidence called on behalf of one party over another has given rise to the question whether adequate reasons for the decision were given: Commissioner for Railways (Qld) v Peters (1991) 24 NSWLR 407 at 415-417 (Kirby P, with whom Priestley JA and Waddell A-JA generally agreed); Federal Commissioner of Taxation v Primary Health Care Ltd [2017] FCAFC 131; 252 FCR 496 at [17] (Kenny, Perram and Robertson JJ). It was held on the facts of one case that the adoption by a judge of one party's submissions, and so acknowledged, is one method of providing adequate reasons: James v Surf Road Nominees Pty Ltd [2004] NSWCA 475 at [168] (Beazley, Tobias and McColl JJA). Other cases concern the question whether, by reference to the decision-maker's reasons, a party's case was in fact considered, which was one of the ways in which the applicant's case was argued here: see, Fletcher Construction Australia Ltd v Lines MacFarlane & Marshall Pty Ltd (No 2) [2002] VSCA 189; 6 VR 1 at [163]-[165] (Charles, Buchanan and Chernov JJA). In C v B [2006] FamCA 513; 35 Fam LR 285, it was submitted to the Full Court of the Family Court of Australia in an appeal by way of rehearing that the reasons of the primary judge were inadequate because they consisted mainly of reproducing, without attribution, the written submissions of counsel for the mother and that they failed to deal with or explain adequately or at all why the submissions for the father were rejected. The court held at [107] that in relation to some issues there was substance in the submission that justice was not seen to be done as a result of the failure of the primary judge to deal with certain submissions advanced on behalf of the father, and the Full Court proceeded to determine the issue for itself. 145 In SZMUV v Minister for Immigration and Citizenship [2009] FCA 205, Flick J expressed reservations about the manner in which a federal magistrate in the decision under appeal expressed his reasons for decision, where four of the nine pages set out verbatim the submissions of the Minister as to the relevant background and the substantive issues, however four of the pages set out the reasons as expressed by the magistrate. The concern that Flick J raised was that it was not self-evident that the magistrate had given independent consideration to the grounds of review that were advanced. On balance, however, Flick J held at [22] that the reasons did disclose a reasoning process involving independent consideration of the case, and displayed no appealable error. A similar conclusion was reached by Flick J in like circumstances in SZNRZ v Minister for Immigration and Citizenship [2010] FCA 107. 146 In Pollard v Wilson [2010] NSWCA 68, pages 8 to 43 of the primary judge's reasons consisted of a verbatim reproduction of submissions made by counsel for one of the parties at trial. The submissions were acknowledged by the primary judge, who expressed agreement with them. On appeal, the court held at [166] that nowhere in the primary judge's reasons did his Honour indicate that he had analysed the relevant evidence and himself concluded that the assertions made by counsel were justified by the evidence, and the court stated that he should have done so. However, the inadequacy of the judge's reasons was held at [176] not to have impeded the court's task on appeal, and therefore was not material to the outcome. 147 LVR (WA) Pty Ltd v Administrative Appeals Tribunal [2012] FCAFC 90; 203 FCR 166 (LVR) was an appeal from the dismissal by a judge of an application for judicial review of a decision of the Tribunal brought under the ADJR Act. The Tribunal's decision was a discretionary decision under s 42A(5)(b) of the AAT Act to dismiss a number of applications on the ground of the appellants' failure to comply with procedural directions. One of the grounds of judicial review was that the Tribunal had failed to take a relevant consideration into account, namely the contents of an affidavit that had been filed, to which the Tribunal had not referred in its reasons. This ground was rejected by the primary judge: LVR (WA) Pty Ltd v Administrative Appeals Tribunal [2011] FCA 1146; 127 ALD 27 at [29] (Gilmour J). On appeal, the Full Court (North, Logan and Robertson JJ) drew to the attention of the parties the apparent extent of verbatim copying by the Tribunal without attribution of the submissions of the second respondent, the Commissioner of Taxation, which had not been addressed by the parties to the appeal. The position of the Commissioner on the appeal was that the appellants had never claimed that the Tribunal did not take account of the affidavit on the ground that the Tribunal's reasons had been copied. The Full Court was critical of the failure of the parties, and in particular the Commissioner, to draw to the attention of the primary judge the fact that material components of the Tribunal's reasons to which the primary judge was taken had been copied from the Commissioner's submissions to the Tribunal, which had not taken account of the affidavit that was in issue. At [43] to [73] the Full Court set out its qualitative assessment of the extent of the Tribunal's copying before concluding at [78] that the Tribunal had failed to take the relevant affidavit into account. At [81] to [112] the Full Court addressed Australian and North American authorities that have considered the legal consequences of copying material into the reasons of a tribunal. Some of the cases concerned the use of "boilerplate" paragraphs, or the use of common information such as country information in migration cases. The perspective that was the subject of the Full Court's consideration was whether cutting and pasting passages into the reasons of the Tribunal was indicative of a failure by the Tribunal to discharge its statutory function: see [91]-[92]. Ultimately, the Full Court concluded at [130] and [136] that the Tribunal's reasons should not be examined as if they were an independent text without reference to their source, and that so understood, the Full Court concluded that the Tribunal did not have regard to the material in the affidavit in question. The consequence of that failure was that the decision of the Tribunal was set aside and the matter referred to the Tribunal for further consideration. 148 An allied issue arose in an appeal where there was wholesale adoption by the Tribunal of views expressed about the appellant by a differently constituted tribunal: MZZZW at [57]. This was held by the Full Court (Tracey, Murphy and Mortimer JJ) to constitute a miscarriage of the Tribunal's task in respect of which there was not necessarily one correct way of characterising the miscarriage. At [30] the court cited authorities that had considered the issue as a denial of procedural fairness, as a constructive failure to exercise jurisdiction, as bias, and as a failure to discharge the Tribunal's statutory function. In the circumstances of that matter, one of the grounds of appeal that was upheld was that, because the Tribunal had copied the reasoning of the earlier Tribunal decision involving the appellant, the Full Court at [77] was not persuaded that the Tribunal member had formed her own independent views about the appellant, which was a requirement of the discharge of the Tribunal's statutory task. 149 In SZRBA v Minister for Immigration and Border Protection [2014] FCAFC 81; 314 ALR 146, the reasons of an independent merits reviewer contained many paragraphs that had appeared in other decisions either in identical terms or very slightly modified. The Full Court (Siopis, Perram and Davies JJ) held that the reviewer had used a method of cutting and pasting earlier decisions to produce his reasons on the appellant's application, which the Full Court held was not surprising, and with which the court expressed sympathy having regard to the volume of similar claims that the reviewer had to consider. In these circumstances, the Full Court held at [21] that the reviewer had accidently overlooked a substantive submission that was put on behalf of the appellant, with the consequence that there had been a denial of procedural fairness. As to a claim of apprehended bias, the Full Court at [35] stated that it saw no reason to depart from the decision in Minister for Immigration and Citizenship v SZQHH [2012] FCAFC 45; 200 FCR 223, which on almost identical facts concluded that apprehended bias was not made out. 150 In Juneja v Tax Practitioners Board [2017] FCA 908; 72 AAR 407, the Tribunal affirmed a decision of the Tax Practitioners Board which refused the renewal of registration of a company as a tax agent on the ground that its director had ceased to be a fit and proper person. One of the grounds of appeal to the court was that by reason of the extent of copying of the respondent's written submissions, the Tribunal constructively failed to exercise its jurisdiction. Besanko J remarked at [86] that there was very little authority on the point that was in issue. His Honour referred to the decision of the Full Court in LVR at [91] as indicating that extensive copying of a party's submissions may lead to the conclusion that there has been a constructive failure by the Tribunal to exercise its jurisdiction, whether that is because the Tribunal has failed to disclose its reasoning or otherwise. Although there was substantial copying of submissions without attribution, at [89] Besanko J was not persuaded that the Tribunal had failed to consider the applicant's case or engage in an active intellectual process in accepting the respondent's submissions. His Honour also noted at [92] that the Tribunal had from time to time in its reasons referred to submissions of the applicants, and did acknowledge that it was repeating or summarising or referring to a submission made by either the applicants or the respondent. His Honour further noted at [93] that the Tribunal, whilst repeating a number of the respondent's submissions, did not simply adopt or copy all of the respondent's submissions. 151 In Rodchompoo v Minister for Immigration and Border Protection [2018] FCA 965, the Tribunal reviewed a decision of a delegate of the Minister who refused to revoke the cancellation of a visa under s 501CA(4) of the Migration Act. The Tribunal affirmed the decision of the delegate. At first instance, McKerracher J found at [45] that there was a substantial similarity between the Minister's statement of facts issues and contentions before the Tribunal, and the Tribunal's reasons. However, at [66] his Honour held that even accepting that there had been a significant amount of replication, there were indications that the Tribunal had specifically considered the relevant evidence. An appeal from this decision was dismissed: Rodchompoo v Minister for Home Affairs [2018] FCAFC 215. 152 In Li v Attorney-General for New South Wales [2019] NSWCA 95; 99 NSWLR 630 the applicant applied to the Supreme Court of New South Wales under the Crimes (Appeal and Review) Act 2001 (NSW) seeking that an inquiry be conducted by a judge of the court with respect to his sentence in respect of convictions for drug-related offences. The application was refused. The application to the judge was not a judicial proceeding, and there was no ability to appeal the decision, but the applicant sought judicial review under s 69 of the Supreme Court Act 1970 (NSW). One of the applicant's claims was that the reasons of the judge for rejecting the application were an edited version of the submissions of counsel for the Attorney-General. 153 At [10], Basten JA formulated the potentially available grounds of review based upon the parties' submissions as follows - (a) adoption of the Attorney's submissions demonstrated that the judge had not in fact formed the relevant state of satisfaction to enable him to dispose of the proceedings; (b) by adopting the Attorney's submissions of 9 March 2018 the judge failed to consider the submissions of the applicant in reply filed on 30 April 2018; (c) to the extent that the Attorney's submissions of 9 March did not fully address the grounds of the application, adoption of those submissions failed to address all of the grounds relied on by the applicant; (d) in the alternative to the foregoing, that is, on the basis that the judge did in fact form a relevant opinion after considering all the relevant material, the reasons were inadequate in that they failed to explain why the applicant's submissions were rejected; and (e) in the further alternative, the adoption of the Attorney's submissions meant that justice was not "seen to be done". 154 Basten JA at [23] held that there was no reason to conclude that the judge did not in fact form the views expressed in relevant paragraphs of the reasons that adopted the Attorney-General's submissions. Material to his Honour's analysis was that although the judge had adopted the Attorney's submissions almost verbatim, at least stylistic amendments were made to every paragraph of the submissions, some paragraphs were omitted, and other brief passages were added. Further, the judge's reference to the applicant's reply submissions, and the quotation of a passage from the reply submissions gave rise to an inference that the judge had read the reply, and Basten JA held at [37] that there was no basis for considering that the judge had failed to consider the reply submissions, and that if he had failed to consider them, then it would have been immaterial because they contained nothing new. 155 At [47], Basten JA addressed cases involving the adoption of the submissions of one party without attribution, stating - However, the lack of attribution is not necessarily a critical factor; although it may indicate a failure to acknowledge what is being done, it will usually be immediately apparent to each party that one party's submissions have been adopted. 156 Material to Basten JA's analysis was his Honour's observation at [48] that the applicant's application for an inquiry was not an adversarial proceeding with conflicting evidence and submissions. Rather, it was an application for an inquiry in the course of which the Attorney-General articulated both the arguments of the applicant, and the responses to those arguments. 157 At [52]-[53], Basten JA held that it had not been demonstrated that any material matters were ignored or not taken into account by the judge, and that the judge was entitled to accept the submission of the Attorney-General. His Honour held that it could not be inferred that the judge did not apply an impartial and independent mind to the issues raised by the application. On the same basis, his Honour also held at [54] that the judge's reasons were not inadequate. Material to that conclusion was that there had not been competing submissions between which the judge had to choose. 158 Basten JA then turned to the question whether there was an independent requirement that "justice be seen to be done". A derivative of this phrase had been employed by Gray J in Sun Alliance Insurance Ltd v Massoud [1989] VR 8 (Sun Alliance) at 18 in the context of adequacy of reasons in judicial proceedings - The adequacy of the reasons will depend upon the circumstances of the case. But the reasons will, in my opinion, be inadequate if: - (a) the appeal court is unable to ascertain the reasoning upon which the decision is based; or (b) justice is not seen to have been done. The two above stated criteria of inadequacy will frequently overlap. If the primary Judge does not sufficiently disclose his or her reasoning, the appeal court is denied the opportunity to detect error and the losing party is denied knowledge of why his or her case was rejected. 159 Basten JA stated at [56] that it was clear that the phrase "justice is not seen to have been done" in Sun Alliance was intended to be a test of the adequacy of reasons, as reference to Gray J's reasons for judgment confirms. A similar reference had been made by McHugh JA in another case concerning adequacy of reasons, Soulemezis v Dudley (Holdings) Pty Ltd (1987) 10 NSWLR 247 at 281. Basten JA stated at [57] that it was doubtful whether this aphorism was ever intended to be a test of the validity of judicial, let alone administrative decision-making, holding at [58] that the aphorism "encapsulates a value or underlying rationale, not an applicable legal rule or legal principle". His Honour noted that there was no reference to any such principle in Public Service Board of New South Wales v Osmond [1986] HCA 7;159 CLR 656, which is the leading authority for the proposition that there is no common law obligation on administrative decision-makers to provide reasons, and stated that the decision in Wingfoot Australia Partners Pty Ltd v Kocak [2013] HCA 43; 252 CLR 480 was inconsistent with any such approach. 160 Basten JA noted at [62] that no claim of reasonable apprehension of bias was pursued before the court, and added that it would not have been sustainable. In conclusion, at [63] his Honour rejected what he described as a "novel approach" based upon the need for "the appearance of justice". 161 At [69], White JA relevantly agreed with Basten JA, and added at [77]-[79] - 77 I accept that a judge who did not bring an independent and impartial mind to bear on an application under s 78 of the Crimes (Appeal and Review) Act would commit jurisdictional error that would render the judge's decision liable to be declared void. But I do not accept that the relevant yardstick is whether there is an appearance that justice has been done by the bringing to bear of an active, impartial and independent judicial mind, nor that it is sufficient to establish jurisdictional error that the substantial adoption of the Attorney's submissions contributed to the creation of an "impression" that the judge did not bring an independent and impartial mind to the determination of the application. Rather, as the Full Court of the Federal Court said in LVR (WA) Pty Ltd v Administrative Appeals Tribunal (2012) 203 FCR 166; [2012] FCAFC 90 at [91]: "the fundamental question is whether there has been a constructive failure on the part of the decision-maker to perform its allotted task." 78 On the present application for judicial review the question is not whether justice has not been seen to be done, but whether the court should be satisfied that the decision-maker did not in fact bring an impartial and independent mind to the decision. That is, the question is not whether Mr Li, or a reasonable person in his position, would have a legitimate sense of grievance that his submissions had not been fully absorbed and analysed and transmuted into a personal conviction on the part of the judge, but whether it should be found as a fact that the judge did not bring his own independent and impartial mind to bear on the issues. 79 There is no reason to doubt that the views expressed by the judge, albeit in terms that reflect the Attorney General's first submissions, did express his own views on the material presented. I agree with Basten JA's reasons (at [19]-[24] and [52]-[53]) that it cannot be inferred from the judge's acceptance of the Attorney's submissions that his Honour did not form the opinions that he expressed. As no clear and credible bases for challenging the sentence were raised, it cannot be inferred that the judge, in accepting almost verbatim the Attorney's submissions that his Honour did not form the opinions that he expressed. As no clear and credible bases for challenging the sentence were raised, it cannot be inferred that the judge, in accepting almost verbatim the Attorney's submissions, did not apply an impartial and independent mind to the issues raised. 162 Brereton JA dissented, holding that - 104 While the adequacy of reasons for a decision is influenced by the circumstances of the case, reasons will be inadequate not only if it is not possible to discern from them the reasoning upon which the decision is based, but also if justice is not seen to have been done.35 As appears from the authorities discussed below, the second limb will be engaged - that is, justice will not be seen to have been done - if the "reasons" are such as to leave the unsuccessful party with a justifiable sense of grievance because they do not appear to be the product of the active application of an independent and impartial mind. (Footnote 35 is a reference to Sun Alliance at 18 (Gray J).) 163 Brereton JA then reviewed a number of authorities, including many of the cases that have considered the legal consequences of a decision-maker adopting the submissions of one party, with or without attribution. At [132], Brereton JA stated - 132 Nonetheless, it is clear that extensive replication, whether or not attributed, of the submissions of one or both parties will not of itself amount to error, so long as the reasons sufficiently reveal that the decision-maker gave independent consideration to the relevant issues. However, this requires that the reasons appear to a reasonable person in the position of the unsuccessful party to demonstrate that the decision-maker has actively engaged with its submissions and contentions, and brought to bear an independent and impartial judgment founded on reason and logic. That requirement will not be satisfied if the reasons leave an impression that arguments have been embraced without serious consideration having been given either to the contrary point of view, or to the application of an independent point of view, as will be the case where reasons which substantially incorporate, reproduce and/or adopt the submissions of one party do not contain indicia, beyond mere formulaic and stylistic changes, that they are the product of the active application of an independent and impartial mind, which has absorbed and analysed the competing submissions to reach a personal conviction on the part of the judge that the orders to be made are just and in accordance with law. 164 Brereton JA then applied these principles to the judge's reasons, which had adopted the Attorney's submissions practically verbatim, and held that by reference to the yardstick that his Honour formulated at [132], they were deficient. His Honour stated at [143]-[144] - 143 In my view, the judge's reasons are, by the relevant yardstick, deficient. The extensive adoption of the Attorney's submissions, not only on uncontroversial matters, but particularly on the formulation and rejection of the applicant's case, with the only reference to the applicant's second submission being to a concession, is such as to create an impression that the judge has not independently and impartially engaged with the applicant's claim and sub- missions and given serious consideration to them. The four matters inserted - which embellish submissions made by the Attorney rather than engage with submissions advanced by the applicant - are insufficient to overcome this impression, and the stylistic changes, and substitution of conclusions for submissions, even less so. A reasonable person in the position of the unsuccessful applicant would have a legitimate sense of grievance that his submissions had not been fully absorbed, analysed and transmuted into a personal conviction on the part of the judge that the orders to be made were just and in accordance with law, and that the judge did not bring his own independent and impartial mind to bear on the issues. The absence of attribution contributes to this impression, because it obscures the nature of what has been done. 144 In my view, therefore, the extent of the unattributed reproduction of the Attorney's submissions, particularly in respect of the formulation of the applicant's case and its disposal, is such as to create an appearance, from the perspective of a reasonable person in the position of the unsuccessful applicant, that his application did not receive proper independent and impartial consideration. Before us, the Attorney conceded as much. Justice was therefore not seen to be done. And as has been explained, in this field, appearances matter, whatever the reality may be. 165 Brereton JA then held at [152]-[154] that the failure to give reasons that satisfied the yardstick that his Honour identified was a constructive failure to exercise jurisdiction with the consequence that, as with a denial of natural justice, the decision was void. 166 In the present case, it is relevant to recognise that the Tribunal's function is administrative, and not judicial. The standard of reasons is therefore that fixed by the AAT Act, either expressly or by implication, and not the standard ordinarily required of a court. The Tribunal's statutory obligation under s 43(2) and (2B) is to give reasons for its decision which are to include its findings on material questions of fact and a reference to the evidence or other material on which those findings were based. It is not necessary for the Tribunal "to refer to every piece of evidence and every contention made by an applicant in its written reasons": WAEE v Minister for Immigration and Multicultural and Indigenous Affairs [2003] FCAFC 184; 236 FCR 593 (WAEE) at [46] (French, Sackville and Hely JJ). What is material is what the Tribunal considers material, and not what is objectively material, because what is required focuses on the subjective thought processes of the Tribunal: Minister for Immigration and Multicultural Affairs v Yusuf [2001] HCA 30; 206 CLR 323 at [65], [68] (McHugh, Gummow and Hayne JJ). The failure by the Tribunal to mention some matter in its reasons may give rise to an inference that the Tribunal did not consider the matter to be material, but not necessarily that the Tribunal did not take the matter into account: see WAEE at [47]. 167 The question whether the Tribunal brought an independent mind to bear upon the review is a factual one, where each case turns on its own circumstances: MZZZW at [43]. Looking at the Tribunal's reasons as a whole I am not persuaded that the adoption without attribution of the respondent's submissions leads to an inference that the Tribunal failed to bring an independent mind to bear upon its review of the respondent's decisions. There are sufficient indications that the Tribunal's reasons are indeed its reasons for decision to which it gave independent consideration. The main indications are: (1) those paragraphs of the Tribunal's reasons to which I referred at [99] above which are independent of the respondent's submissions, which include summaries of the legal principles, reference to the applicant's submissions with some comments upon them by the Tribunal, the framing of issues that arose on the review, and some conclusions; (2) the fact that there was some material re-ordering of the paragraphs that were reproduced from the respondent's submissions which further supports the application of independent consideration; and (3) the Tribunal's express statement at [62] of its reasons that it had undertaken a thorough analysis of the documentary evidence and had considered the oral testimony. On the face of the Tribunal's reasons, there are insufficient reasons to conclude that the Tribunal did not do what it said it had done. All of this is in a statutory context where the respondent, as the decision-maker, had an obligation under s 33(1AA) of the AAT Act to use its best endeavours to assist the Tribunal to make its decision, which as a corollary must involve the Tribunal acting upon that assistance where appropriate. Without more, the Tribunal does not abrogate its independence by accepting, including to the point of adopting, the submissions of the primary decision maker, which is a party to the proceeding, is entitled to be heard, and which moreover has a statutory obligation to assist the Tribunal to make its decision.

(2) Does the unattributed copying of the respondent's submissions by the Tribunal give rise to an inference that any particular matter that the Tribunal was required to consider was overlooked? 168 Before I address the specific issues relied on by the applicant in support of the grounds relating to Question 6 to which I referred at [130] above, I will make some further observations. 169 It was a theme of many of the submissions advanced on behalf of the applicant that the errors that it claimed occurred were the product of the Tribunal's adoption of the respondent's written submissions. This had an air of the post hoc ergo propter hoc fallacy. If it be assumed that the Tribunal's reasons are its reasons for its decision, it does not necessarily follow that there is a logical connection between the errors that the applicant alleged, and the adoption of the respondent's submissions in material parts of the Tribunal's reasons. On the other hand, there may be cases, of which LVR is an example, where an inference that a tribunal failed to consider a matter that it was required to consider is supported by the fact that the submissions of one party were copied, and which made no reference to the relevant matter. 170 Throughout the applicant's written submissions to the court there were claims that the Tribunal had failed to take into account relevant considerations, citing Minister for Aboriginal Affairs v Peko-Wallsend Ltd [1986] HCA 40; 162 CLR 24. The applicant characterised various features of its arguments relating to facts and evidence that must be taken to have been rejected by the Tribunal, as amounting to relevant considerations. 171 Generally speaking, the Tribunal is required, as an incident of its obligation to accord procedural fairness, to hear and consider the parties' submissions as a whole. For the purposes of judicial review on the ground of error of law, the Tribunal may err if it fails to do so. The Tribunal may also err if it fails to take account of some mandatory consideration, that is, a consideration to which the governing statute requires the Tribunal to have regard, either expressly or by implication. A mandatory consideration may, in a particular case be "a substantial, clearly articulated argument relying upon established facts": Dranichnikov v Minister for Immigration and Multicultural Affairs [2003] HCA 26; 214 CLR 496 at [24] (Gummow and Callinan JJ). The failure to address a claim of this type may amount to a failure to accord procedural fairness, or a constructive failure to exercise jurisdiction. However, not every argument about questions of fact in issue will amount to a matter requiring separate consideration: Re Minister for Immigration and Multicultural Affairs; Ex parte Durairajasingham [2000] HCA 1; 74 ALJR 405 (Durairajasingham) at [36] (McHugh J); Minister for Immigration and Citizenship v SZRKT [2013] FCA 317; 212 FCR 99 at [97] (Robertson J), cited in Minister for Immigration and Border Protection v MZYTS [2013] FCAFC 114; 230 FCR 431 at [68]-[70] (Kenny, Griffiths and Mortimer JJ). A Tribunal is not required "to give a line-by-line refutation of the evidence for the claimant either generally or in those respects where there is evidence that is contrary to findings of material fact made by the Tribunal": Durairajasingham at [65] (McHugh J); see also, Coal of Queensland Pty Ltd v Innovation and Science Australia at [144] (Logan, Griffiths and Moshinsky JJ). Accordingly, the failure to address specifically every argument about the evidence will not necessarily support an inference that an applicant's arguments were not considered: WAEE at [46]; Steed v Minister for Immigration and Ethnic Affairs (1981) 37 ALR 620 at 621 (Fox, McGregor and Morling JJ). 172 In this case, many of the submissions advanced on behalf of the applicant to the court complained of the failure of the Tribunal to address factual matters that the applicant claimed refuted the Tribunal's findings. The flaws with these types of arguments include that they were advanced on at least the implicit basis that the Tribunal should have accepted the factual premises underpinning them, and that they inevitably invite the court to review the merits of the Tribunal's decision: see, Abebe v Commonwealth [1999] HCA 14; 197 CLR 510 at [195] (Gummow and Hayne JJ). 173 The final observations are directed to some submissions made on behalf of the applicant that the Tribunal failed to inform itself of information necessary to arrive at the correct decision, including information that was not before the Tribunal, and on which the applicant now seeks to rely. Although the Tribunal's function is inquisitorial, and is supported by statutory powers to get information, the AAT Act expressly contemplates that directions may be made, as they were in this case, for procedures to be followed in connection with the proceeding, including requiring a party to provide a statement of matters or contentions: AAT Act, s 33(1A), (2) and (2A). The decision-making function of the Tribunal in the present case was to conduct a review under the AAT Act where, pursuant to directions that were made, statements of facts issues and contentions and statements of evidence were filed by the parties, where a public hearing took place, and where the applicant and the respondent had the opportunity to open their cases, to examine witnesses, and to make submissions. In a review of this kind, it was for the applicant corporation to advance whatever evidence or argument it wished to advance: see, Abebe v Commonwealth at [187] (Gummow and Hayne JJ). In undertaking its review function the Tribunal is generally entitled to be guided by the issues that the parties choose to put before it for consideration: Repatriation Commission v Morris (1997) 79 FCR 455 at 460 (Beaumont J); Paramananthan v Minister for Immigration and Multicultural Affairs (1998) 94 FCR 28 at 63 (Merkel J). Although the Tribunal may make its own inquiries and may require the production of further material, the duty of the Tribunal to conduct a review and to arrive at the correct or preferable decision does not equate to a duty to make inquiries or to get information. In particular, the Tribunal here was not required to construct a review case for the applicant: DCR19 v Minister for Immigration, Citizenship, Migrant Services and Multicultural Affairs [2021] FCAFC 229 at [80] (Bromwich, Anastassiou and Anderson JJ), citing Abebe v Commonwealth at [187]. However, there may be exceptional cases where a failure to make an obvious inquiry about a critical fact may amount to a constructive failure to review: see, Minister for Immigration and Citizenship v SZIAI [2009] HCA 39; 259 ALR 429 at [18], [25] (French CJ, Gummow, Hayne, Crennan, Kiefel, and Bell JJ), and see also Prasad v Minister for Immigration and Ethnic Affairs [1985] FCA 46; 6 FCR 155 at 169-170 (Wilcox J), and Minister for Immigration and Citizenship v Le [2007] FCA 1318; 164 FCR 151 at [60] (Kenny J). In addressing the applicant's submissions I also bear in mind the following statement in Enichem Anic Srl v Anti-Dumping Authority (1992) 39 FCR 458 at 469 (Hill J, with whom Gummow J and O'Connor J agreed) - Decision-making is a function of the real world. A decision-maker is not bound to investigate each avenue that may be suggested to him by a party interested. Ultimately, a decision-maker must do the best on the material available after giving interested parties the right to be heard on the question. 174 See also, Hossain v Minister for Immigration and Border Protection [2018] HCA 34; 264 CLR 123 at [28] (Kiefel CJ, Gageler and Keane JJ), and MZAPC v Minister for Immigration and Border Protection [2021] HCA 17; 390 ALR 590 at [32] (Kiefel CJ, Gageler, Keane and Gleeson JJ), citing the first sentence of the above passage. 175 I now turn to consider the specific submissions that the applicant raised which relied on the Tribunal's substantial reproduction of the respondent's submissions in its reasons in support of claims that various matters were not considered by the Tribunal. Many of the submissions which are considered below were also relied on in connection with Question 4, to which I will return.

(a) The 2016 R&D material 176 The Tribunal had determined at the directions hearings on 14 September and 23 October 2018 that the 2016 year did not form part of the review before the Tribunal, and confirmed this when making a ruling upon evidence at the commencement of the hearing. I do not consider that this course was a product of the adoption of the respondent's closing submissions, or was indicative of any failure thereby to discharge the Tribunal's function of independent review. I will later consider more generally the applicant's claim that the Tribunal erred in determining not to include the 2016 year as part of the review.

(b) The applicant's written closing submissions in reply 177 The Tribunal referred to the applicant's reply submissions at [60] and [61] of its reasons which I have set out at [113] and [114] above. At [60], the Tribunal extracted a quotation from the reply and described two of the reply's main features, namely a submission that the respondent had failed to substantiate its case, and the applicant's claim that there were discrepancies in the evidence relied on by the respondent. At [61], the Tribunal extracted another quotation from the reply, and identified five arguments about the evidence which emerged from four sub-paragraphs, including by adding a page span and exhibit reference which were not included in the reply for one of the pieces of evidence to which the applicant had referred. During the hearing before the court, the applicant's representative submitted that no part of [61] was reasonably attributable to the reply submission, on the basis that there was no reference to the arguments or submissions put forward. However, after considering the reply for myself, I must reject this submission, because it is plain that at [61] the Tribunal did summarise some of the main points advanced in the reply. Although the applicant's reply submissions to the Tribunal made detailed references to the evidence, the statutory standard of reasons did not require that the Tribunal set out and address all these references. The Tribunal having referred to the reply submissions in the way it did, there is no reason to find that the Tribunal did not read and consider the reply submissions in the course of analysing the documentary and oral evidence.

(c) The purpose and objective of the applicant's R&D activities 178 The applicant submitted that the Tribunal did not take into account the purpose and object of its R&D activities, which it submitted was to develop consumer information technology products. It was submitted that the purpose of early-stage testing was not to determine things such as whether eating causes weight gain, but to develop and confirm concepts and algorithms necessary to produce the product that the applicant was developing. It was submitted that the Tribunal failed to consider that experiments in the 2014 and 2015 years were carried out at the early stages of the development of an innovative product, and that the Tribunal wrongly impugned this R&D as if it were required to produce new physiology science. In partial support of this claim, the applicant submitted that in deciding that the testing was inadequate for not having a sufficiently large number of test participants and not having female test participants, the Tribunal failed to consider that the type of R&D activity appropriate to an information technology product depends on the stage in the product development process, that the early testing in the 2014 and 2015 years was necessary and suited to that stage, and that the applicant was carrying out work necessary to develop systems, algorithms, and concepts to produce the product planned. These submissions raise issues which overlap with Question 2, which I will address later, that the Tribunal failed to consider the totality of the product being developed. 179 I do not accept these submissions, which seek to transmute arguments about evaluation of the evidence to mandatory relevant considerations that the Tribunal was bound to consider. Section 27J of the IR&D Act requires the making of findings "about an R&D entity's registration under section 27A for an income year". The Tribunal's approach was to consider the whole of the evidence against the activities that the applicant had registered for the two relevant years, and to find that there was insufficient evidence of what was actually done by the applicant or by Akyman that would support a finding that the applicant had engaged in "core R&D activities" or "supporting R&D activities". Further, the Tribunal did address the applicant's purpose at [80], [88], and [97] of its reasons, and made findings adverse to the applicant in relation to the requirement in s 355-25(1)(b) that the experimental activities be conducted for the purpose of generating new knowledge.

(d) The contemporaneous evidence of the applicant's R&D activities 180 In its written submissions to the court, the applicant submitted that "the Tribunal erred in adopting the Respondent's submissions that there was not contemporaneous evidence of registered core R&D activities", and that "[t]he Tribunal did not take into account that the different iterations of functional specifications and detailed design documents, and other documents, were contemporaneous evidence of the 2014 and 2015 R&D activities". This submission was tantamount to alleging that the Tribunal erred by not finding that there was contemporaneous evidence of the 2014 and 2015 activities. Self-evidently, this submission goes to the merits of the Tribunal's decision, and does not raise a question of law: see, Abebe v Commonwealth at [195] (Gummow and Hayne JJ). 181 These submissions led to a further submission that, "[t]he Tribunal failed to inform itself of information necessary to come to a correct decision, and refused to consider relevant material the Applicant wished it to consider". In support of this further submission, the applicant submitted that "[h]ad the Tribunal not made this refusal, it would properly have considered the information that is in evidence in the current Tribunal proceeding 2018/6110 concerning the 2016 year", and referred to the evidence before the court of material before the Tribunal in that application which included further witness statements, and an expert report of a Dr Paul Nash dated 28 July 2020 that was prepared more than 12 months after the Tribunal's decision the subject of this appeal. These further submissions, raised under the guise of error by the Tribunal in failing to consider the contemporaneous evidence of the R&D activities, were premised on a claim that the Tribunal had a duty to inform itself of information necessary to come to a correct decision, and had refused to consider relevant material that the applicant wished it to consider. 182 As to the claimed obligation of the Tribunal to inform itself of relevant information, I have referred at [173]-[174] above to the relevant principles. There are no features of this review which obliged the Tribunal to venture outside the voluminous material that was put before it by the parties, and to make its own inquiries. Still less was there any obligation on the Tribunal to seek to advance the applicant's case on its behalf, or to advise the applicant that the evidence that it had put before the Tribunal was insufficient to establish its claims, which is an issue that I will consider further when addressing Question 5. 183 As to the claimed failure to consider relevant information, this relates to the Tribunal's decisions at the directions hearings on 14 September 2018 and 23 October 2018 to proceed with the hearing that had been fixed to commence on 29 October 2018 in relation to the review of the respondent's findings relating to 2014 and 2015 years, together with other alleged failures to consider material relating to the 2016 year which I will separately address when considering Question 1.

(e) Circumstances that were favourable to the reliability of the evidence of the applicant's witnesses 184 The applicant submitted that the Tribunal erred in making its findings about the credibility of the applicant's witnesses at [69] of its statement of reasons, which I have set out at [116] above. As I have noted elsewhere, those findings were reproduced from the respondent's written submissions to the Tribunal. The applicant submitted that by copying examples of contradictory, confused or evasive evidence from the respondent's submissions, the Tribunal "considered irrelevant rather than relevant considerations in order to impugn the witness' testimony". The applicant submitted that there were other features of the evidence that the Tribunal did not consider. 185 In relation to the credibility findings, the applicant cited the decision of the Full Court in MZZZW, to which I referred at [148] above. In MZZZW, the appellant alleged jurisdictional error by the Refugee Review Tribunal in affirming a decision of a delegate of the Minister to refuse a protection visa. The decision of the Tribunal that was challenged was the second decision of the Tribunal on the appellant's application for review, an earlier decision having been quashed. The second Tribunal, differently constituted, again affirmed the decision of the delegate. Many passages of the second Tribunal's reasons were identical to passages in the decision record of the first Tribunal. The passages that were copied included passages dealing with findings on credibility in circumstances where the appellant had appeared before the first and the second Tribunal. A troubling aspect of the second Tribunal's reasons in MZZZW was that there was reference to evidence that was before the first Tribunal, and findings that purported to be the second Tribunal's own findings based upon that evidence, which were actually copied from the findings of the first Tribunal. At [43], and with reference to the decision of French J in WAFK v Minister for Immigration and Multicultural and Indigenous Affairs [2003] FCA 1293; 133 FCR 209, the Full Court characterised the copying of passages concerning credibility as "an abdication of the Tribunal's task". At [44], the Full Court distinguished WAFK from the case before it on the ground that in the former, what was copied "was not prior opinions of another Tribunal member about the individual claims of the same applicant". At [57], the Full Court referred to the non-delegable nature of the function conferred on the particular member of the Tribunal undertaking the review function. The Full Court stated that the task "is not to be performed by adopting the views of a differently constituted tribunal about the same applicant", and then added - Whatever the criticism might be of a decision-maker transferring across findings by another decision-maker about a different applicant and applying them to the applicant before the decision-maker, those criticisms pale against the wholesale adoption of the opinions of a previously constituted tribunal about the same applicant. 186 As I mentioned earlier, in these circumstances the Full Court in MZZZW was not satisfied that the second Tribunal brought an independent mind to bear upon the review of the appellant's claims. 187 In my view, the Full Court's decision in MZZZW is distinguishable from the present case. In MZZZW, the adoption without attribution of the reasons and findings of the earlier decision-maker, when the appellant had appeared before both Tribunals, had special significance in supporting an inference that the Tribunal had not conducted an independent review. Similar conclusions may arise where a decision-maker upon internal review repeats word for word substantial portions of the reasons of the first decision-maker: Huluba v Minister for Immigration and Ethnic Affairs (1995) 59 FCR 518. This case is different, because for reasons that I have given I am not persuaded on the whole of the evidence that the Tribunal did not conduct an independent review. The findings of the Tribunal, although adopted substantially from the respondent's written submissions, were findings based upon the evidence before the Tribunal, and not some other decision-maker. Amongst other things, in the face of the Tribunal's statement at [62] of its reasons that it had conducted a thorough analysis of the documentary evidence together with a consideration of the oral testimony, I am not persuaded to find that the Tribunal's findings, including those as to credibility, were not the Tribunal's own findings as a result of its own independent review of the documentary evidence before it, and the oral evidence that it heard. 188 To the extent that the applicant's submissions maintained that the Tribunal failed to address relevant considerations that would support the credibility of the applicant's witnesses, I do not accept those submissions because they invite merits-based review by the court in relation to which I refer to the observations that I made at [170]-[172] above.

(f) Evidence of a Dr Jessie Hiu Kiu that was favourable to the applicant 189 The applicant submitted that because the Tribunal adopted the submissions of the respondent, it had failed to consider the "evidence" of Dr Jessie Hiu Kiu that was favourable to the applicant. The evidence before the Tribunal would support a finding that in 2015 Dr Kiu was a representative of "AusIndustry" who met with Mr Mark Nicolau and Mr Werner Nicolau and prepared a report following the meeting. Dr Kiu did not give evidence before the Tribunal, but she was referred to in other evidence. Further, it is not apparent that Dr Kiu was the decision-maker in respect of the reviewable decision, which was made by a delegate of the respondent board. 190 In his second witness statement before the Tribunal dated 17 May 2018, and to which I referred at [77] above, Mr Mark Nicolau referred to a meeting that had taken place on 4 June 2015 with representatives of "AusIndustry" which he and Mr Werner Nicolau had attended. The meeting was also referred to in a document before the Tribunal dated 9 July 2015 titled "R&D Tax Incentive Compliance Review" which was signed by Dr Kiu and another representative, Alison Sare, who was described as the Acting Deputy State Manager. A note of a telephone conference involving representatives of the Australian Taxation Office on 26 May 2016 was also in evidence before the Tribunal which referred to another meeting between representatives of the applicant and "AusIndustry" on 17 May 2016. Dr Kiu is not referred to in that document. Within that note were a number of pithy lines, including "Having amply demonstrated new knowledge is being developed". In its written closing submissions to the Tribunal, the applicant relied on select extracts from the above documents in support of its claims. 191 I do not accept the applicant's submissions that the Tribunal failed to consider evidence of Dr Kiu that was favourable to the applicant. For one thing, the submission assumes that the "evidence" that was referred to was favourable to the applicant, when it was well open to the Tribunal to conclude that the documents relied on were either not material to the Tribunal's own review on the material that was before the Tribunal, or otherwise did not support the applicant's claims. As I stated earlier at [141], the Tribunal's function was to "do over again" what was done by the original decision-maker upon the material before the Tribunal. 192 The applicant cited the reasons of Kirby J in Shi v Migration Agents Registration Authority [2008] HCA 31; 235 CLR 286 at [37]-[38] in support of its submission that the Tribunal had erred in failing to consider the "evidence" of Dr Kiu. However, the passages cited by the applicant do not support a submission that there was any error of law by the Tribunal in this respect. At [37], Kirby J referred to a decision of Davies J sitting as the Tribunal in Re Control Investment Pty Ltd and Australian Broadcasting Tribunal [No 2] (1981) 3 ALD 88 at 92-93, stating - Davies J acknowledged that regard might be had to the decision of the primary decision-maker as part of the "material before the Tribunal", particularly where it involved special expertise or knowledge. But ultimately, it was for the Tribunal to reach its own decision upon the relevant material including any new, fresh, additional or different material that had been received by the Tribunal as relevant to its decision. In effect, this was no more than a consequence of the Tribunal's obligation to conduct a true merits review. (Footnotes omitted.) 193 While the Tribunal might in a particular case have regard to the reasons of a primary decision-maker, the above passage does not support a submission that it was obliged to do so. Instead, Kirby J emphasised that it was for the Tribunal to reach its own decision, for which there is abundant authority. Further, I have already noted above that it is not apparent that Dr Kiu was the primary decision-maker. Moreover, [74]-[75] of the Tribunal's reasons show that the Tribunal did not overlook the applicant's submissions in relation to statements attributed to Dr Kiu, albeit that these passages adopted the respondent's submissions. Ultimately, the applicant's submissions in relation to statements in documents attributed to Dr Kiu and officers of the Australian Taxation Office go no further than to raise factual matters which the Tribunal considered.

(g) Evidence that the applicant had carried out testing 194 The applicant submitted that as a consequence of copying the respondent's submissions as its own reasons the Tribunal failed to consider that there was evidence that the applicant had carried out testing, and that product specifications and other documents that the applicant produced were a product of that testing. In support of its written submission in relation to this claim, the applicant cited the expert report of Dr Nash to which I referred at [181] above, and which was prepared more than 12 months after the Tribunal's decision. The applicant's reliance on Dr Nash's report in this way was outside the limited terms on which leave was given to adduce that evidence. Further, the reliance on Dr Nash's report in this way only serves to emphasise that the applicant was seeking to cavil with the merits of the Tribunal's decision. As with other submissions of the applicant, this submission rises no higher than a claim that the Tribunal did not accept evidence relating to factual matters that were before the Tribunal. 195 With reference to its claim that the Tribunal failed to take into account evidence that the applicant had conducted testing, the applicant repeated its submissions that "the Tribunal failed to inform itself of the information necessary to come to a correct decision, and refused to consider the relevant information the Applicant wished the Tribunal to consider" to which I referred at [181], and referred to the evidence that it has filed with the Tribunal in the separate proceeding relating to the review of the respondent's findings in respect of the 2016 year. I will address these claims when considering Question 1.

(h) Claimed limitations on the expertise of the respondent's witnesses 196 The applicant claimed that in copying the respondent's submissions as its reasons the Tribunal did not take into account the expertise and testimony of the witnesses in respect of the applicant's development of algorithms. This was in essence a complaint that the Tribunal had at [82] and [87] of its reasons placed weight on Dr Kerr's evidence that she did not recognise diagrams that were in evidence as algorithms in circumstances where her principal expertise was as a dietician, and where the applicant claimed that Dr Schneider had accepted that the applicant had developed algorithms. The applicant made similar submissions to the Tribunal in its written closing submissions. As with other submissions of the applicant, the flaw in this submission is that it is premised on factual assumptions. The applicant did not refer to the whole of Dr Schneider's evidence. Whether Dr Schneider accepted that the applicant had developed algorithms was a question of fact for the Tribunal, as the following evidence of Dr Schneider in cross-examination before the Tribunal at T225/30-37 demonstrates - Q: In your first report you stated there is no evidence that an algorithm was developed to a level of detail suitable for implementation? A: Yes, I stand with that statement. Q: So someone having those diagrammatic representations of the algorithms, cannot write a piece of code, and representing the functionality of the algorithm? A: If I had to take those diagrams alone, and no additional information, I would not be able to implement those algorithms. 197 For the foregoing reasons, the matters relied on by the applicant in its submissions alleging specific errors as a result of the Tribunal's reproduction of the respondent's submissions in its statement of reasons do not disclose any error of law by the Tribunal and do not, either alone or in combination, otherwise give rise to an inference that there was any relevant matter not taken into account by the Tribunal.

(3) Is there a requirement that administrative justice must appear to have been done? 198 I will now turn to the question whether by adopting substantially the submissions of the respondent as its reasons the Tribunal erred by failing to discharge its statutory review function in a way that would promote public trust and confidence in the decision-making of the Tribunal. I stated at [134] above that another way of expressing this question is to ask whether there is a requirement that administrative justice must appear to have been done. The issue that I will consider is whether, having rejected any claim that it is to be inferred that the Tribunal in fact failed to undertake an independent review, was the Tribunal in breach of an additional requirement that it should appear to some objective observer to have done so? The applicant did not raise as a ground of appeal that the Tribunal's decision was affected by a reasonable apprehension of bias, and did not advance any submissions to this effect. In the course of the hearing I specifically drew attention to the fact that apprehended bias had not been raised. A finding of apprehended bias should not be made lightly: CNY17 v Minister for Immigration and Border Protection [2019] HCA 50; 268 CLR 76 at [56] (Nettle and Gordon JJ), citing Re JRL; ex parte CJL [1986] HCA 39; 161 CLR 342 at 371 (Dawson J). A corollary is that any allegation of apprehended bias should be squarely put. In the way that the applicant's submissions were advanced, any claim of apprehended bias would have had to confront the observations of Gummow A-CJ, Hayne, Crennan and Bell JJ in Michael Wilson & Partners Ltd v Nicholls [2011] HCA 48; 244 CLR 427 at [67], referring to the fallacious argument that apprehended bias in the nature of prejudgment might be demonstrated by reasons that do not deal with all of the losing side's arguments. As no allegation of apprehended bias was raised or argued in this case, the question whether the reasonable lay observer might have thought that the Tribunal might not have brought an independent mind to the review was not a question that was before the court for determination. Rather, one of the ways in which the case was addressed by the applicant was to claim that the Tribunal failed to carry out its statutory functions in a way that would promote public trust and confidence in the decision-making of the Tribunal (picking up the language of s 2A(d) of the AAT Act), and to submit that the Tribunal's decision "could not give a reasonable observer confidence [that] the Tribunal fairly dealt with, or considered the applicant's case". 199 Any requirement along the lines that "administrative justice must be seen to have been done" must be found within the AAT Act, as either an express or implied incident of the lawful exercise of the Tribunal's review function. As to express provisions, s 2A(d) of the AAT Act is aspirational or exhortatory in nature, and is not a source of directly enforceable rights and obligations: Fard v Secretary, Department of Immigration and Border Protection [2016] FCA 417 at [80] (Griffiths J), citing Minister for Immigration and Multicultural Affairs v Eshetu [1999] HCA 21; 197 CLR 611 at [108] (Gummow J); Twentyman v Secretary, Department of Social Services [2018] FCA 1892; 163 ALD 517 at [79] (Wigney J); Doan v Minister for Home Affairs [2019] FCA 1172 at [37], [49] (Griffiths J). However, I would not understand the characterisation of s 2A of the AAT Act as being aspirational or exhortatory to deny the relevance of s 2A to the question whether, in a particular case, the Tribunal failed to act in accordance with some implied condition of the conferral of its functions such as the condition that it exercise those functions reasonably: see, Minister for Immigration and Citizenship v Li [2013] HCA 18; 249 CLR 332 at [96]-[98] (Gageler J). 200 That brings me to the question whether there is some implied requirement that the Tribunal must discharge its functions in a way that gives a reasonable person confidence in the Tribunal's decision-making. In Li v Attorney-General for New South Wales, to which I referred at [152]-[165], Brereton JA in dissenting reasons at [132] adopted as a yardstick that reasons for a decision must appear to a "reasonable person in the position of the unsuccessful party" to demonstrate that the decision-maker has actively engaged with its submissions and contentions, and brought to bear an independent and impartial judgment founded on reason and logic. This yardstick is concerned with imputed lay perception rather than whether in fact it is to be inferred that the decision-maker made some specific error. However, the majority comprising Basten JA and White JA rejected the notion that in administrative decision-making there was any requirement that justice must be seen to be done. There was no submission to me that the majority reasoning in Li v Attorney-General for New South Wales was clearly wrong, and in all the circumstances there is no reason why I should not apply it. To the analyses of Basten JA and White JA I add my own observation that there are established principles relating to the implied condition that the Tribunal exercise its review function free of a reasonable apprehension of bias. The centrepiece of those principles is the hypothetical fair-minded lay observer. In circumstances such as the present case, where there is no claim of apprehended bias, where there is no sufficient reason to conclude that the Tribunal did not in fact consider the applicant's claims, and where there is no sufficient reason to conclude that the Tribunal did not in fact form the views set out in its reasons, it would lead to incoherence to introduce a further principle which conditions the lawful exercise of jurisdiction on the perceptions of the hypothetical "reasonable person in the position of the unsuccessful party".

Question 1 - activities in the 2016 year 201 Question 1 is in the following terms - In the circumstances, including that: a. the Respondent had denied that the Applicant conducted registered R&D activities in the 2014 and 2015 years; b. the Applicant contended to the Tribunal that the completion of the 2014 and 2015 years' registered R&D activities was proven by, amongst other things, the evidence of the Applicant's R&D accomplishments in the subsequent income year ended 30 June 2016 ("2016 year"), because the Applicant's 2016 year's accomplishments could not have been made without the completion of the 2014 and 2015 years' registered R&D activities; c. the Tribunal refused to consider the 2016 year's R&D accomplishments on the basis that this was "irrelevant to the Tribunal's review" in respect of the 2014 and 2015 years which were the subject of the proceeding, did the Tribunal make an error of law by refusing to admit evidence of the Applicant's 2016 year's R&D activities, or by failing to take into account relevant considerations, including the relevant consideration of the Applicant's R&D accomplishments in the 2016 year, or by acting with Wednesbury unreasonableness? 202 There was some overlap between the issues raised by Question 1, and the issues raised by Question 5. It is convenient to deal with the overlapping issues together here. The applicant submitted that its R&D activities for the 2014 and 2015 years were part of a multi-year project, that the applicant had carried out conceptual design, functional specifications, detailed design and test planning in the 2014 and 2015 years, and in the 2016 year conducted testing and prototyping. The applicant submitted that it had sought to submit evidence of its 2016 R&D activities, as these were probative and relevant to the matter in dispute, and that "the Tribunal refused to admit evidence of the 2016 year's R&D activities which proved the Applicant conducted the 2014 and 2015 years' registered R&D activities". In its written submissions before the court the applicant identified the evidence filed with the Tribunal in proceeding 6110 of 2018 as containing "the evidence the Applicant sought to provide to the Tribunal in this matter". Specifically, it was submitted that the evidence was the expert report of Dr Paul Nash dated 28 July 2020 to which I referred at [181] above, and two affidavits of Mr Mark Nicolau dated respectively 12 May 2020 and 26 May 2020 and filed in the subsequent proceeding in the Tribunal. Both affidavits contain indorsements indicating that they were prepared by solicitors acting for the applicant. The report of Dr Nash and the two affidavits of Mr Mark Nicolau were part of the evidence that I received on this appeal on the limited basis to which I referred at [65], namely that its use was confined to the identification of what evidence might have been available to the applicant to place before the Tribunal but for the alleged errors of law that it raised by its notice of appeal. The applicant claimed that the documents evidenced that the 2016 year's R&D activities and accomplishments were derived from and relied upon the activities carried out in the 2014 and 2015 years, and that they evidenced that the 2014 and 2015 core R&D activities took place. 203 The applicant submitted that the legal consequences of its claim that the Tribunal had refused to admit this evidence were that the Tribunal had acted unreasonably, and that it had failed to accord procedural fairness to the applicant. Specifically, the applicant submitted that the Tribunal's exercise of power "was so unreasonable no reasonable Tribunal could have precluded relevant evidence and found against the Applicant on the basis of insufficient evidence", citing Minister for Immigration and Citizenship v Li at [24], [26], [63]-[76], [77]-[81], [82]-[83], [85], and Re Refugee Review Tribunal; Ex parte Aala [2000] HCA 57; 204 CLR 82 at [101], [103]. 204 The respondent submitted that the rulings made by the Tribunal to proceed with the hearing were not the subject of this appeal. I accept that the rulings themselves are not decisions for the purposes of s 44 of the AAT Act, because they were not the effective decision of the Tribunal on the application for review: Director General of Social Services v Chaney [1980] FCA 87; 47 FLR 80 at 103-104 (Deane J, Fisher J agreeing). However, it is open to allege error of law in the effective decision of the Tribunal by reference to some anterior interlocutory decision that affects the final order: The News Corporation Ltd v National Companies and Securities Commission (1984) 5 FCR 88 at 95 (Fox J), 103 (Woodward J) and 122 (Beaumont J); cf, in relation to judicial decisions, Gerlach v Clifton Bricks Pty Ltd [2002] HCA 22; 209 CLR 478 at [6] (Gaudron, McHugh and Hayne JJ). If the Tribunal's decisions to proceed with the hearing resulted in a denial of procedural fairness, or were otherwise legally unreasonable and material to the outcome of the review, then they can be said to affect the final order of the Tribunal which is the subject of appeal under s 44. 205 I have summarised the applicant's submission in the terms in which it was put, namely that the Tribunal "refused to admit evidence". However, save for one respect to which I refer in the next paragraph, the applicant's submission does not reflect what took place. The proceeding before the Tribunal had been prepared for hearing and related only to the respondent's findings for the 2014 and 2015 years, and not the 2016 year. The preparation included the parties' statements of facts issues and contentions, statements of lay evidence, and expert reports. The initial hearing dates of 16 and 17 May 2018 had been vacated so as to allow the applicant further time to file reply evidence, and the hearing of the matter was refixed to commence on 29 October 2018. The applicant's application to the Tribunal on 14 September 2018 was to have the Tribunal include in the proceeding a review of the respondent's findings for the 2016 year. That application faced the difficulty that there was no valid application to the Tribunal in respect of the reviewable decision of 31 October 2017 relating to the 2016 year. Further, what would likely have been involved if the applicant's application had been acted upon by the Tribunal was the vacation of the hearing date, and the preparation of further statements of facts issues and contentions, lay witness statements, and expert evidence. 206 At the hearing itself, the Tribunal ruled as inadmissible the brief statement of Mr Werner Nicolau dated 7 August 2018 to which I referred at [56] above on the ground that it "purely deals with the '15/'16 year". The substance of the statement went no further than to state that (at the time of making the statement) Mr Nicolau had not received a response from the respondent to its request to review its decision in respect of the 2016 year. There was no material error of law by the Tribunal in refusing to admit that statement. In the course of the hearing the Tribunal also treated as irrelevant other material, including material relating to sockets in walls, which Mr Mark Nicolau appeared to accept had nothing to do with the proceeding. The Tribunal also stated in the course of the hearing that the subject-matter of a "ZAP contact app" was irrelevant on the ground that its development was not the subject of the applicant's claim for the taxation incentive for the relevant years. On other occasions, during the course of his cross-examination of Dr Kerr and Dr Schneider, Mr Mark Nicolau accepted that a cloud security system was not relevant to the years in question because it related to the 2016 year. 207 The applicant submitted that the Tribunal's decisions to proceed with the hearing of the review was legally unreasonable. One of the matters that has been held to inform the legal standard of reasonableness is whether the subject-matter of the power is substantive or procedural: Minister for Immigration and Border Protection v Stretton [2016] FCAFC 11; 237 FCR 1 at [71] (Griffiths J, Allsop CJ and Wigney J agreeing). Here, the Tribunal's discretionary decisions to proceed with the hearing that had been fixed to commence on 29 October 2018 in relation to the review of the respondent's findings for the 2014 and 2015 years were procedural in nature. However, although procedural, the decisions related to the process by which the Tribunal's decision was made, that is, the manner in which the Tribunal discharged its review function. The High Court has held often enough that the implied obligation that a statutory decision-maker discharge its function rationally and reasonably applies also to the process of decision-making. 208 The threshold of legal unreasonableness is usually high: Minister For Home Affairs v DUA16 [2020] HCA 46; 385 ALR 212 (DUA16) at [26] (Kiefel CJ, Bell, Keane, Gordon and Edelman JJ). In Minister for Immigration and Citizenship v Li, Gageler J at [98] referred in the following terms to the standard of legal unreasonableness in relation to a procedural decision of the Migration Review Tribunal that was subject to exhortations and aspirations in the Migration Act - … the manner of its performance of a procedural duty or of its exercise or non-exercise of a procedural power is so unreasonable that no reasonable tribunal heeding those exhortations or adhering to those aspirations could have done what the [Tribunal] in fact did … 209 Also reflecting the high threshold of legal unreasonableness, is the statement of Hayne, Kiefel and Bell JJ in Minister for Immigration and Citizenship v Li at [67] that the question to which the standard of reasonableness is addressed is "whether the statutory power has been abused". 210 Whether the Tribunal's decisions to proceed with the hearing in relation to the 2014 and 2015 years were legally unreasonable is to be determined at the time those decisions were made, and turns on the circumstances that were before the Tribunal at those times: DUA16 at [26]. The applicant placed much emphasis on the evidence that it subsequently marshalled in relation to its separate application for review of the respondent's decision in respect of the 2016 year. But the evidence in that form was not in existence, still less before the Tribunal in September and October 2018 at the time the Tribunal confirmed that hearing fixed to commence on 29 October 2018 would proceed. 211 The exhortations in s 2A(b) and (c) of the AAT Act are that the Tribunal provide a mechanism of review that "is fair, just, economical, informal and quick", and "is proportionate to the importance and complexity of the matter". It may be accepted that the applicant might have considered that it was forensically advantageous to have the Tribunal review the respondent's findings for the 2014 and 2015 years together with those for the 2016 year. It may also be accepted that there might have been some advantages in having all issues determined together, as the applicant had submitted to the Tribunal. On the other hand, acceptance of the applicant's submissions to the Tribunal would have involved further delay and expense in circumstances where there was no valid application then on foot to review the respondent's findings in respect of the 2016 year. Consistently with the exhortations in s 2A of the AAT Act, it was within the range of reasonable outcomes for the Tribunal to determine to proceed with the hearing that had been fixed, and for which the parties had prepared their statements of facts issues and contentions and their evidence. The applicant has not demonstrated that the Tribunal's rulings were arbitrary, irrational, or otherwise unreasonable to the requisite threshold. 212 Reasonableness is closely linked to procedural fairness: Minister for Immigration and Citizenship v Li at [92] (Gageler J). A refusal to grant an adjournment can constitute a failure to give a party the opportunity of adequately presenting its case. However, the relevant duty of the Tribunal was to give the applicant a reasonable opportunity to present its case, and "neither the Act nor the common law imposes upon the Tribunal the impossible task of ensuring that a party takes the best advantage of the opportunity to which [it] is entitled": Sullivan v Department of Transport (1978) 20 ALR 323 at 343 (Deane J, Fisher J agreeing). 213 In my view, the applicant was given a reasonable opportunity to present its case. The application for review proceeded on the accepted basis that it concerned the respondent's findings for the 2014 and 2015 years, and the applicant had the opportunity to present its evidence and to make submissions in relation to the Tribunal's review of those findings. The material relied on by the applicant before the Tribunal included material relating to the 2016 year. Activities conducted in a later year may be capable of shedding appropriate light on the characteristics of the activities conducted in the years in question: Coal of Queensland Pty Ltd v Innovation and Science Australia at [131] (Logan, Griffiths and Moshinsky JJ). Whether they did so was a question of fact. In this case, the relevance of the applicant's claimed activities in 2016 to the activities that the applicant claimed it undertook in 2014 and 2015 was in issue before the Tribunal. The relevance of the applicant's material relating to the activities such as the development of a "ZAP contact mobile app", a "loyalty card system", an "integrated payment gateway", a "find my phone app", and a "transaction card processing system" to the years in question was the subject of cross-examination of Mr Werner Nicolau and Mr Mark Nicolau at the hearing. Documents that were said to relate to such activities were admitted into evidence subject to weight. The applicant submitted to the Tribunal in its written closing submissions that "information on the Health and Fitness System relating to the year ended 30 June 2016 is probative of the nature of the R&D project into the Health and Fitness System that was being developed in 2014 and 2015", and relied on activities in the 2016 year in its written reply submissions. The Tribunal evidently did not accept these submissions, and held at [71]-[72], set out under [117] above, that much of material relied on by the applicant relating to its registered activities for the 2016 year was irrelevant to the Tribunal's review on the ground that the activities were not mentioned in the applicant's registrations for the relevant years. The Tribunal's findings did not constitute a failure by the Tribunal to consider this material in the sense that the Tribunal did not address the material. Rather, for the reasons the Tribunal gave it considered the material to be irrelevant to its determination, which involved a factual finding that raises no question of law. 214 As to the applicant's submissions in support of Question 1 that in refusing to admit evidence of the applicant's R&D activities for the 2016 year the Tribunal "failed to take into account relevant considerations", including by failing to "have regard to the best information available", I reject the submissions for three reasons. The first is that for the reasons I have given what occurred is not accurately characterised as a refusal to admit relevant evidence. The second reason relates to that part of the applicant's submissions in support of Question 1 that claimed that as an incident of its inquisitorial function the Tribunal itself was required to seek out adequate material in order to arrive at the correct or preferable decision. I have dealt with those submissions at [173], and for the reasons set out I do not accept them. The third reason is that to the extent that the applicant claimed that the Tribunal erred at [71] and [72] of its reasons by concluding that material relating to the applicant's 2016 activities was irrelevant to its review, and that the Tribunal thereby failed to have regard to relevant considerations, the claim invites the court to engage in merits-based review: see [172] and [213] above.

Question 2 - the totality of the product 215 Question 2 is in the following terms - Did the Tribunal make an error of law by failing to take into account relevant considerations for the purpose of applying section 355-25 of the Tax Act, or by acting with Wednesbury unreasonableness, including in failing to consider the totality of the product being developed and thereby denying that the necessary precursor R&D activities in the individual registration years were part of the overall broader R&D activities necessary to develop the total product? 216 The applicant submitted that the Tribunal erred by failing to consider the whole R&D project in determining whether the registered core R&D activities in each year satisfied the statutory requirements, and submitted that in this respect the Tribunal failed to consider relevant evidence. The applicant further submitted that, had the Tribunal considered that totality of the product, then even if the Tribunal came to the view that the activities for 2014 and 2015 were not core R&D activities, the Tribunal "would have been sufficiently aware of the project development to conclude that those activities were necessary precursors to the 2016 year's registered R&D activities, and could therefore be characterised as supporting R&D activities to the 2016 year's core R&D activities, which involved experimentation and validation of the specifications and detailed designs of the R&D activities performed in the 2014 and 2015 years, and development of several support mobile applications". The applicant relied on s 27J(1)(c)(iii) of the IR&D Act, which I have set out at [18] above, to submit that it had been open to the respondent to find that a registered core R&D activity was a supporting R&D activity in relation to one or more core R&D activities "yet to be conducted for which the entity could be registered in the registration year if those activities were conducted during the registration year". The applicant submitted that the Tribunal acted unreasonably in failing to consider the totality of its health and fitness system, and failed to take into account relevant considerations, "to the extent that no reasonable tribunal could have so acted". 217 In response, the respondent submitted that the Tribunal had considered the applicant's material in relation to its activities for the 2016 year, and at [71]-[72] had found the material to be irrelevant. The respondent submitted that the applicant had contended only shortly before the hearing before the Tribunal that its activities for the 2016 year were relevant to its registered activities for the 2014 and 2015 years, and did so without any elaboration or factual support. The respondent also submitted that the applicant had adopted inconsistent positions before the Tribunal, because in its reply dated 17 May 2018 to the respondent's statement of facts issues and contentions the applicant had submitted only that the respondent's findings in respect of the 2016 year were "probative in relation to the Respondent's process of decision making to make the findings in the 2013/14 and 2014/2015 years", and that the applicant had acknowledged in its reply statement that "the Applicant's research and development activities were different in the 2015/2016 income year from the 2013/2014 income year and the 2014/2015 income year": see the extract at [69] above. The respondent further submitted that no argument had ever been advanced by the applicant to the Tribunal that the applicant's registered activities for the 2014 and 2015 years were supporting activities on the basis that that they were directly related to core activities claimed for the 2016 year (reflecting the definition of supporting R&D Activities in s 355-30 of the ITAA), and submitted that it was legally reasonable for the Tribunal to make its decision without considering a peripheral argument that had not been put to it. 218 The main difficulty with the applicant's claims and submissions directed to Question 2 is that after stating that it had analysed all the oral and documentary evidence, the Tribunal was just not satisfied that any experimental activity of the type described by the applicant in the registrations was carried out: see the Tribunal's reasons at [78], [85], [92], [96] and [97]. Amongst other things, in the Tribunal's estimation this was a product of insufficient direct evidence of what was actually done, the conclusory nature of the evidence in the applicant's witness statements, and the unsatisfactory nature of the documentary evidence: see the Tribunal's reasons at [66] and [70]. I do not accept that the Tribunal failed to consider the evidence before it, as distinct from finding that much of the material on which the applicant had relied was irrelevant to its determination, which was a factual issue that was contested before the Tribunal. As with other submissions and claims of the applicant, there is an underlying premise here that the material that the Tribunal regarded as irrelevant was in fact probative and supportive of the applicant's claims, and was therefore a "relevant consideration". This premise involves the evaluation of questions of fact which were for the Tribunal, and invites merits review: see [172] above. 219 As to the applicant's submission that if the Tribunal had considered the totality of the product it might have considered that the registered activities for the 2014 and 2015 years were supporting activities in respect of the 2016 year, there are a number of problems with this claim. The first is that the applicant did not put that submission to the Tribunal. The Tribunal was not called upon to consider whether the registered core R&D activities for the 2014 and 2015 years were supporting R&D activities for the 2016 year, which would have required consideration of whether the 2014 and 2015 activities were directly related to one or more core R&D activities: ITAA, s 355-30(1). In all the documents that the applicant filed with the Tribunal, this claim was not made. In both Moreton Resources Ltd v Innovation and Science Australia and Coal of Queensland Pty Ltd v Innovation and Science Australia it was acknowledged, consistently with the terms of s 27J(1)(c)(iii) of the IR&D Act, that registered core R&D activities may be found to be supporting R&D activities in respect of one or more specified core R&D activities yet to be conducted and which could be registered. However, in both cases it is clear that submissions to that effect had been made to the Tribunal: see Moreton Resources at [163], and Coal of Queensland at [89], [148], [151]-[152]. The second problem is that the submission does not overcome the Tribunal's findings that, on the evidence, it was not satisfied that the claimed activities occurred. The third problem is that the claim is not fairly within the questions of law raised by the applicant on appeal to the court.

Question 3 - construction of s 355-25(1)(a) of the ITAA 1997 220 Question 3 is in the following terms - Did the Tribunal make an error of law by applying the wrong test to the Applicant's R&D activities, in that the Tribunal incorrectly construed section 355-25(1) of the Tax Act, which states that core R&D activities are experimental activities "whose outcome cannot be known or determined in advance", to mean "there must be a high level of uncertainty as to whether the outcome can be known or determined in advance" (at [39] of the Tribunal's decision), causing the Tribunal to conclude that the Applicant's experimental activities were ineligible as their outcome could have been known in advance, under that wrong test? 221 Section 355-25 of the ITAA 1997 is set out at [11] above. At [37] of its reasons, the Tribunal accepted the respondent's submission that that the condition in s 355-25(1) of the ITAA 1997 that a core R&D activity be "an experimental activity whose outcome cannot be known or determined in advance on the basis of current knowledge, information or experience" required that "there must be a high level of uncertainty as to whether the outcome can be known or determined in advance". And at [39] of its reasons, the Tribunal accepted the respondent's submission that s 355-25 requires that there is, objectively, "a high degree of uncertainty about the outcome of the activities, and that the risk must be scientific or technical in nature". 222 The applicant submitted that the Tribunal erred by substituting for the text of s 355-25(1) of the ITAA 1997 a formulation that had a significantly different effect. The applicant submitted that the Tribunal's error in construction caused it to conclude that the applicant's experimental activities were ineligible. The applicant made other submissions which asserted factual matters, such as that "[o]nly the R&D activities conducted by the Applicant could have achieved the specific knowledge required to implement the Health and Fitness System". 223 The respondent submitted that the construction of s 355-25(1) adopted by the Tribunal was supported by the text and the context of the provision. As to the text, the respondent pointed to the juxtaposition of the phrases "cannot be" and "but can only be determined" (emphasis added) in s 355-25(1)(a). The respondent cited a passage from the reasons for judgment of Windeyer J in Council of the Municipality of Randwick v Rutledge [1959] HCA 63; 102 CLR 54 at 94, where his Honour stated that "[t]he presence of "exclusively", "solely", or "only" always adds emphasis and is not to be disregarded", citing R v Cockburn (1852) 16 QB 480 at 491. The respondent also relied on the objects of Division 355 and extrinsic material, in particular the explanatory memorandum to the Bill which led to the enactment of the Tax Laws Amendment (Research and Development) Act 2011 (Cth), which inserted Division 355 of the ITAA 1997. In relation to core R&D activities, the explanatory memorandum provided, inter alia - … 2.12 To qualify as core R&D activities, experiments will not merely confirm what is already known, or have an outcome that can be known or determined in advance. Rather, they will be activities whose outcome can only be determined by employing a systematic progression of work based on scientific principles and using an approach that proceeds from hypothesis to experiment, observation and evaluation and leads to logical conclusions. This approach is generally known as the scientific method. [Schedule 1, item 1, paragraph 355-25(1)(a)] 2.13 The requirement for the scientific method establishes a threshold for the knowledge gap and the degree of uncertainty that an eligible experiment must seek to address. The threshold will not be met if the knowledge of whether something is scientifically or technologically possible, or how to achieve it in practice, is deducible by a competent professional in the field on the basis of current knowledge, information or experience. 2.14 Further, the nature of the eligible experiment is such that there will be a clear risk that the outcome of the experiment will not be the desired one. The potential for this risk to deter firms from undertaking knowledge-generating R&D underpins the rationale for the R&D tax incentive. 2.15 Less rigorous knowledge discovery and problem solving techniques, such as 'trial and error' alone, will not be sufficient to qualify as eligible experimentation. However, trial and error may form part of an eligible experiment where the conditions for core R&D are met. Trial and error can also qualify as a supporting activity where it forms part of a decision to proceed to activities that qualify as core R&D. … 2.18 The need to employ the scientific method also reflects the degree of novelty in the ideas being tested. That is, the knowledge being sought must go beyond validating a simple progression from what is already known and beyond merely implementing existing knowledge in a different context or location. Rather, the gap between existing knowledge and the hypothesis being investigated will be significant enough to require application of the scientific method. … 224 The respondent submitted that the Tribunal's statement at [39] of its reasons that s 355-25(1)(a) requires "objectively, a high level of uncertainty about the outcome of the activities" is consistent with the text, context and extrinsic materials, and submitted that the applicant had not advanced any alternative construction. The respondent further submitted that, moreover, even if the Tribunal did misconstrue s 355-25(1)(a) of the ITAA 1997, it was not material to the Tribunal's conclusions that the applicant's registered activities were ineligible. That was because the principal reason for the Tribunal's conclusion that the applicant's registered activities were ineligible was that the Tribunal was not satisfied that the activities as registered were conducted in the relevant years. The applicant submitted in addition that at [80], [88], and [97] of its reasons the Tribunal had also held that the requirements in s 355-25(1)(b) of the ITAA 1997 had not been satisfied, being the requirement that there be experimental activities that are conducted for the purpose of generating new knowledge (including new knowledge in the form of new or improved materials, products, devices, processes or services). 225 I am not persuaded that the Tribunal misconstrued s 355-25(1)(a) of the ITAA 1997 in any way that was material to the outcome of the review. It is important to read the whole of [36] to [41] of the Tribunal's reasons together: see the extracts at [107] above. The Tribunal was correct to observe at [37] that the phrase "cannot be" in s 355-25(1)(a) is absolute. To equate this with "a high level of uncertainty" is to introduce a gloss on the statutory language that is not found in the extrinsic material or any authority. I am not persuaded that it was necessary to introduce a gloss to explain s 355-25(1)(a), but I do not think the gloss was material. If anything, a "high level of uncertainty" is a slight retreat from the absolute terms of the text of the provision. Further, the first limb of s 355-25, which is a negative stipulation, is not to be read in isolation but is to be read in conjunction with the positive condition in the second limb that the outcome of the experimental activities "can only be determined by applying a systematic progression of work" based upon principles of established science and the staged processes referred to in s 355-25(1)(a)(ii). I do not consider that there was any material error at [39] of the Tribunal's reasons in drawing together the conditions of s 355-25(1)(a) and concluding that they indicated a requirement that objectively there had to be a high degree of uncertainty about the outcome of experimental activities. 226 Moreover, I accept the submissions of the respondent that even if the Tribunal had misconstrued s 355-25(1)(a) of the ITAA 1997, it was not material to the outcome of the review for the fundamental reason that owing to what the Tribunal considered was the unsatisfactory state of the evidence, the Tribunal was just not satisfied that any experimental activity of the type described by the applicant in its registrations was carried out: see [218] above. Further, I also accept the submission of the respondent that the Tribunal was not satisfied that the separate and necessary condition in s 355-25(1)(b) had been engaged, that is, at [80], [88], and [97] of its reasons the Tribunal was not satisfied that in relation to the registered activities the applicant had a purpose of generating new knowledge.

Question 4 - miscellaneous claims of error 227 Question 4 is in the following terms - Did the Tribunal make an error of law by taking irrelevant considerations into account, or by not taking relevant considerations into account, or by acting with want of jurisdiction, or acting in excess of jurisdiction, including as follows: a. The Tribunal, in determining if the experimental activities qualified as core R&D activities, did not take into account the relevant consideration of the purpose and objective of the experimental activities, which was to scientifically develop algorithms for information technology products that could be used by a consumer. The Tribunal wrongly impugned the experimental activities directed towards making new pioneering algorithms and information technology products and that applied existing scientific advances in physiology made by others, for not producing those scientific advances in physiology made by others. b. The Tribunal, in deciding that there was not contemporaneous evidence of registered core R&D activities, did not take into account the relevant consideration that the different iterations of the functional specifications and detailed design documents were contemporaneous evidence of registered R&D activities in the relevant years. c. The Tribunal, in impugning the experimental activities on the basis of inconsistencies between the oral testimony of the Applicant and the oral testimony of the Applicant's research service provider, did not take into account the relevant consideration that the research service provider, and not the Applicant, had carried out the experimental activities and had detailed knowledge of the experimental activities that the Applicant did not. d. The Tribunal failed to take into account the relevant consideration of the evidence of Dr Jessie Hiu Kiu that was favourable to the Applicant's case, only taking into consideration the evidence of Dr Jessie Hiu Kiu that could go against the Applicant's case and reading Dr Jessie Hiu Kiu's assessment solely as negative to the Applicant. e. The Tribunal, in deciding that the Applicant had not carried out testing, failed to take into account the relevant consideration that there was evidence that the Applicant had carried out testing, including desktop testing, and that the product specifications and documents were a product of that testing. f. The Tribunal, in deciding that the testing was inadequate for not having a sufficiently large number of test participants and not having female test participants, failed to take into account the relevant consideration that the type of R&D activity that is appropriate to an information technology product depends on the stage at which the product is in the product development process; the early testing in relation to the Health & Fitness System product was suited to the early stage of the product development process. g. The Tribunal accepted the Respondent's expert witness Dr Kerr's view that the applicant had not developed algorithms and did not take into account the relevant consideration that the Respondent's software engineering expert, Dr Schneider had accepted that the Applicant had developed algorithms, whereas Dr Kerr is a dietician and is not an expert in algorithms. 228 The applicant's submissions in support of its claims relating to Question 4 included submissions that the claimed errors were a product of the Tribunal's adoption of the respondent's submissions. In considering for the purposes of Question 6 whether the Tribunal's adoption of the respondent's submissions resulted in any specific error, I have already addressed most of the submissions made in relation to Question 4: see [175]. 229 All of the claims raised by Question 4 are rejected on the global basis that they do not raise any questions of law, but seek to cavil with factual findings made by the Tribunal on the alleged ground that the Tribunal failed to accept evidence that is claimed to have been favourable to the applicant's case and which is wrongly characterised by the applicant as "relevant considerations". I have already referred to and explained this flaw in the applicant's approach to the appeal at [122], [170]-[172], [179], [188], and [196].

Question 5 - denial of procedural fairness 230 Question 5 is in the following terms - Did the Tribunal, in a proceeding involving an Applicant without legal representation, make an error of law, including in: a. not informing the Applicant at the hearing, nor at any stage before making its decision, of the Tribunal's assessment that the Applicant's documentary evidence was insufficient; b. not seeking relevant evidence from the Applicant's research service provider in respect of matters for which documentary evidence was perceived by the Tribunal to be deficient; c. refusing to consider the Applicant's R&D activities and accomplishments in the 2016 year, which the Applicant contended could not have been achieved without the completion of those registered R&D activities in question in the 2014 and 2015 years; d. declining to review the Respondent's reviewable decision in respect of the Applicant's registered R&D activities in the 2016 year concurrently with the review of the 2014 and 2015 years; e. then making a decision against the Applicant on the basis that "insufficient evidence exists to indicate that any of the registered core R&D activities were conducted" (at [62] of the Tribunal's decision), thus failing to afford the Applicant the fairness required under section 2A of the Administrative Appeals Tribunal Act 1975 ("AAT Act"), procedural fairness and a fair trial? 231 The components of Question 5 raise two issues. The first issue is whether the Tribunal erred by failing to warn the applicant that its evidence was insufficient, and by failing to seek further evidence. The second issue is whether there was an error of law by the Tribunal in refusing to consider the applicant's claimed R&D activities for the 2016 year. 232 As to the first issue, the Tribunal was not required to give the applicant a running appraisal of its claims, or to warn the applicant that it might not succeed. It should have been obvious to the applicant from the respondent's submissions to the Tribunal and from the conduct of the hearing that its claims were contested, particularly on the ground that the applicant's evidence was insufficient, which was specifically raised at [39] of the respondent's statement of facts issues and contentions that was filed and served months before the hearing. Further, the expert reports filed and served on behalf of the respondent months before the hearing gave the applicant ample notice of claimed deficiencies in its material, and the applicant had a reasonable opportunity to respond to that material. There was no requirement on the Tribunal as an incident of its obligation to accord procedural fairness to state what was obvious: see, Minister for Immigration and Citizenship v SZGUR [2011] HCA 1; 241 CLR 594 at [9] (French CJ and Kiefel J, Heydon J and Crennan J agreeing), citing Commissioner for Australian Capital Territory Revenue v Alphaone Pty Ltd [1994] FCA 1074; 49 FCR 576 at 591-2. The fact that the applicant corporation was not represented at the hearing before the Tribunal by legal practitioners does not affect this conclusion. 233 The applicant submitted that the hearing was conducted in an inappropriate and adversarial manner, characterised by frequent interjections from the Tribunal which hindered the applicant's presentation of its case. The applicant also submitted that the Tribunal "did not seek the required evidence" from the applicant's witnesses. 234 In footnote (81) of the applicant's written submissions to the court, the applicant referred to 41 transcript references in support of these submissions. I have read all of these references, and they do not support the applicant's submission that the Tribunal acted inappropriately. The transcript reveals that the Tribunal questioned the relevance of some of the questions put by Mr Mark Nicolau in his cross-examination of witnesses. The Tribunal intervened to ensure that documents to which the cross-examiner referred a witness were properly identified and available. The Tribunal dealt with a number of objections to questions when objections were raised on behalf of the respondent. On a number of occasions the Tribunal was concerned to have what appeared to be comments by the cross-examiner formulated into questions for the witness. Other questions were considered by the Tribunal to be too lengthy, and the Tribunal required that they be re-formulated. On some occasions, the Tribunal intervened to require that a witness be allowed to complete an answer to a question. The Tribunal at one point sought to have Mr Mark Nicolau clarify some evidence that he gave in cross-examination that the Tribunal evidently thought was inconsistent with other evidence. As I mentioned earlier, at one point the Tribunal allowed Mr Werner Nicolau to continue the cross-examination of Dr Kerr, which had been commenced by Mr Mark Nicolau. Yet this was relied on by the applicant as one of its transcript references to support a submission that the Tribunal hindered the applicant in the presentation of its case. 235 I am left with the impression that the applicant included in footnote (81) of its submissions transcript references to statements made by the Tribunal during the course of evidence, without regard to their context or content, or the occasion for the Tribunal's intervention. What the applicant relied on as hindering interjections can be reasonably viewed as part of the normal course of a hearing process, and included assistance provided by the Tribunal to the applicant's representatives. Further, while the Tribunal assisted the representatives of the applicant from time to time by, for instance, encouraging them to frame questions properly, and to focus on the review as a fresh determination, rather than one concerned with the identification of error in the decision under review, the Tribunal was not obliged to make its own inquiries, or to conduct the applicant's case on its behalf with a view to improving its evidentiary foundation. The applicant was given a reasonable opportunity to prepare its evidence, and there were no special circumstances of this case that made it legally unreasonable for the Tribunal not to make some inquiry of its own, still less to undertake some large-scale investigation into the applicant's claims so as to improve their evidentiary basis. Rather, the position was the conventional position described by Hill J in Enichem Anic Srl v Anti-Dumping Authority which I cited at [173] above, namely that the Tribunal had to do its best on the material available. 236 In relation to the second issue, the applicant submitted that its R&D activities for the 2014 and 2015 years were part of a multi-year project, that the applicant had carried out conceptual design, functional specifications, detailed design and test planning in the 2014 and 2015 years, and in the 2016 year conducted testing and prototyping. The applicant submitted that "the Tribunal refused to admit evidence of the 2016 year's R&D activities which proved the Applicant conducted the 2014 and 2015 years' registered R&D activities". In support of that submission, the applicant relied on what occurred at the directions hearings on 14 September 2018 and 23 October 2018 where the Tribunal determined that the findings for the 2016 year were not part of the review in circumstances where no valid application to review the respondent's findings in respect of the 2016 year had been filed. I have addressed those submissions in the course of considering Question 1, and for the reasons I have given, I do not accept them.

Conclusions 237 The appeal will be dismissed. I certify that the preceding two hundred and thirty-seven (237) numbered paragraphs are a true copy of the Reasons for Judgment of the Honourable Justice Wheelahan.

Associate: Dated: 24 May 2022 SCHEDULE GROUNDS OF APPEAL 1. The Respondent made an incorrect original decision in denying that the Applicant had conducted any of the registered core and supporting R&D activities in the 2014 and 2015 years. The Applicant is entitled to have the decision set aside, and to obtain R&D tax incentives in respect of the Applicant's expenditure on registered R&D activities, on the basis that: a. The Applicant conducted R&D activities that were registered, and were core R&D activities within the meaning of section 355-25 of the Tax Act and supporting R&D activities within the meaning of section 355-30 of the Tax Act, in the 2014 and 2015 years. b. The Applicant conducted experimental activities that were required to develop new pioneering algorithms and information technology products, whose outcome could not be known or determined in advance on the basis of current knowledge, information or experience, but could only be determined by applying a systematic progression of work that was based on principles of established science and proceeding from hypothesis to experiment, observation and evaluation, and leading to logical conclusions, that were conducted for the purpose of generating new knowledge in the form of new products, devices, processes and services. c. There is contemporaneous documentary evidence proving the R&D activities were carried out, and there is corroborative evidence showing that the Applicant's 2016 year's accomplishments could not have been made without the completion of the 2014 and 2015 years' registered R&D activities. 2. The Tribunal incorrectly affirmed the Respondent's decision, as a consequence of the following conduct that amounts to errors of law. a. The Tribunal made an error of law by refusing to admit the Applicant's evidence of the 2016 year's R&D accomplishments. b. The Tribunal made an error of law by failing to take into account relevant considerations, including the relevant consideration of the Applicant's R&D accomplishments in the 2016 year. c. The Tribunal made an error of law by acting with Wednesbury unreasonableness in failing to take into account relevant considerations, including the relevant and probative consideration of the Applicant's R&D accomplishments in the 2016 year. d. The Tribunal made an error of law by failing to take into account relevant considerations for the purpose of applying section 355-25 of the Tax Act, including in failing to consider the totality of the product being developed and thereby denying that the necessary precursor R&D activities in the individual registration years were part of the overall broader R&D activities necessary to develop the total product. e. The Tribunal made an error of law by acting with Wednesbury unreasonableness, including in failing to consider the totality of the product being developed and thereby denying that the necessary precursor R&D activities in the individual registration years were part of the overall broader R&D activities necessary to develop the total product. f. The Tribunal made an error of law by applying the wrong test to the Applicant's R&D activities, in that the Tribunal incorrectly construed section 355-25(1) of the Tax Act, which states that core R&D activities were experimental activities "whose outcome cannot be known or determined in advance", to mean "there must be a high level of uncertainty as to whether the outcome can be known or determined in advance" (at [39] of the Tribunal's decision), causing the Tribunal to conclude that the Applicant's experimental activities were ineligible as their outcome could have been known in advance, under that wrong test. g. The Tribunal made an error of law in determining if the experimental activities qualified as core R&D activities, by not taking into account the relevant consideration of the purpose and objective of the experimental activities, which was to scientifically develop algorithms for information technology products that could be used by a consumer. The Tribunal wrongly impugned the experimental activities directed towards making new pioneering algorithms and information technology products, and that applied existing scientific advances in physiology made by others, for not producing those scientific advances in physiology produced by others. h. The Tribunal made an error of law in deciding that there was not contemporaneous evidence of registered core R&D activities, as it did not take into account the relevant consideration that the different iterations of the functional specifications and detailed design documents were contemporaneous evidence of registered R&D activities in the 2014 and 2015 years. i. The Tribunal made an error of law in impugning the experimental activities on the basis of inconsistencies between the oral testimony of the Applicant and the oral testimony of the Applicant's research service provider, as it did not take into account the relevant consideration that the research service provider, and not the Applicant, had carried out the experimental activities and had detailed knowledge of the experimental activities that the Applicant did not. j. The Tribunal made an error of law in failing to take into account the relevant consideration of the evidence of Dr Jessie Hiu Kiu that was favourable to the Applicant's case, only taking into consideration the evidence of Dr Jessie Hiu Kiu that could go against the Applicant's case, and reading Dr Jessie Hiu Kiu's assessment solely as negative to the Applicant. k. The Tribunal made an error of law in deciding that the Applicant had not carried out testing, in that the Tribunal failed to take into account the relevant consideration that there was evidence that the Applicant had carried out testing, including desktop testing, and that the product specifications and documents were a product of that testing. l. The Tribunal made an error of law in deciding that the testing was inadequate for not have a sufficiently large number of test participants and not having female test participants, in that it failed to take into account the relevant consideration that the type of R&D activity that is appropriate to an information technology product depends on the stage at which the product is in the product development process; the early testing in relation to the Health & Fitness System product was suited to the early stage of the product development process. m. The Tribunal made an error of law in that it accepted the Respondent's expert witness Dr Kerr's view that the applicant had not developed algorithms and did not take into account the relevant consideration that the Respondent's software engineering expert, Dr Schneider had accepted that the Applicant had developed algorithms, whereas Dr Kerr is a dietician and is not an expert in algorithms. n. The Tribunal made an error of law in a proceeding involving an Applicant without legal representation by failing to afford the Applicant procedural fairness, the fairness required under section 2A of the AAT Act, and a fair trial, including in declining to review the Respondent's reviewable decision in respect of the Applicant's registered R&D activities in the 2016 year concurrently with the review of the 2014 and 2015 years. o. The Tribunal made an error of law in a proceeding involving an Applicant without legal representation by failing to afford the Applicant procedural fairness, the fairness required under section 2A of the AAT Act, and a fair trial, including in not informing the Applicant at the hearing, nor at any stage before making its decision, of the Tribunal's assessment that the Applicant's documentary evidence was insufficient. p. The Tribunal made an error of law in a proceeding involving an Applicant without legal representation by failing to afford the Applicant procedural fairness, the fairness required under section 2A of the AAT Act, and a fair trial, including in not seeking relevant evidence from the Applicant's research service provider in respect of matters for which documentary evidence was perceived by the Tribunal to be deficient. q. The Tribunal made an error of law in a proceeding involving an Applicant without legal representation by failing to afford the Applicant procedural fairness, the fairness required under section 2A of the AAT Act, and a fair trial, including in refusing to consider the Applicant's R&D activities and accomplishments in the 2016 year, which the Applicant contended could not have been achieved without the completion of those registered R&D activities in question in the 2014 and 2015 years. r. In substantially reproducing the Respondent's written submissions, without attribution, as the Tribunal's reasons for decision, the Tribunal made errors of law in: (i) failing to exercise its jurisdiction; (ii) failing to exercise independent judgment; (iii) acting unreasonably; (iv) failing to take into account relevant considerations; (v) taking into account irrelevant considerations; and (vi) failing to carry out its statutory functions in a way that would promote public trust and confidence in the decision-making of the Tribunal.