Introduction
This paper, which is an update of the Parliamentary
Library’s November
2023 publication, provides a summary of how research and development
(R&D) and innovation expenditure is measured in Australia. It is intended
to help readers find R&D and innovation data and to assist with
understanding what the common metrics mean and the differences between them.
This paper concludes with Table 1, which
provides a quick summary of terminology, definitions and sources.
For an overview of the current state of R&D and
innovation investment in Australia, including a summary of the latest key
statistics and recent policy developments in R&D and innovation, please refer
to the Parliamentary Library’s 2025 research paper, R&D and innovation in Australia: 2024 update.
How is
investment in R&D measured?
First released in 1963, the Organisation for Economic
Co-operation and Development’s (OECD) Frascati
manual: guidelines for collecting and reporting data on research and
experimental development (Frascati
Manual) has long provided a ‘common language for talking about R&D
[research and experimental development] and its outcomes’.[1] The seventh, and
most recent, edition of the Frascati Manual was released in 2015 and provides
an internationally recognised methodology for collecting and using R&D
statistics.
The Frascati Manual defines R&D as follows:
[Research and experimental
development (R&D)] comprise creative and systematic work undertaken in
order to increase the stock of knowledge – including knowledge of humankind,
culture and society – and to devise new applications of available knowledge.[2]
The Frascati Manual adds that for an activity to be
considered an R&D activity it must be ‘novel, creative, uncertain,
systematic, [and] transferable and/or reproducible’.[3] R&D applies to
3 types of activities:
- basic
research – which is ‘experimental or theoretical work undertaken primarily to
acquire new knowledge … without any particular application or use in view’
- applied
research – which also aims to acquire new knowledge but with a ‘specific,
practical aim or objective’
- experimental
development – which is systematic work drawing on previous research which aims
at ‘producing new products or processes or to improving existing products or
processes’.[4]
The manual describes 2 different, but complementary,
ways of measuring how much governments invest in R&D called the funder‑based
and performer‑based approaches.[5]
The funder‑based approach measures the funding sources
of R&D and is the approach used in the annual science, research and
innovation budget tables (SRI Budget Tables), which are discussed further
below. In Australia, the key funder‑based measurements are the government
budget allocations for R&D (GBARD) and the funding of industry R&D tax
measures (currently the R&D
Tax Incentive). The combination of these figures is referred to as the
government’s ‘total R&D investment’ in the SRI Budget Tables, or ‘total
government support of R&D’ by the OECD.
The performer‑based approach considers what funds were
used by different sectors for R&D activities. The ABS obtains performer‑based
data by surveying organisations to elicit information on their intramural
expenditure on R&D.[6]
The ABS survey data is then used in the calculation of R&D expenditure in
the business (BERD), government (GOVERD), higher education (HERD), and private
non-profit (PNPERD) sectors of the economy.
As noted in the Frascati Manual, there are likely to be
differences between the values calculated using these 2 approaches for a
range of reasons, including sampling difficulties and tracking/reporting
differences.[7]
As a result, it is generally not appropriate to directly compare funder‑based
and performer‑based figures.
How
is innovation measured?
First published in 1992, the OECD’s Oslo
manual: guidelines for collecting, reporting and using data on innovation
(Oslo Manual) is the equivalent to the Frascati Manual in the field of
innovation measurement. The most recent edition of the Oslo Manual (the fourth
edition published in 2018) defines innovation as follows:
An innovation is a new or
improved product or process (or combination thereof) that differs significantly
from the unit’s previous products or processes and that has been made available
to potential users (product) or brought into use by the unit (process).[8]
The Oslo Manual notes that this definition relates to
innovation as an outcome, but the term can also be used to describe a process.
The manual uses the term ‘innovation activities’ to describe
innovation-as-a-process, which it defines as ‘all developmental, financial and
commercial activities undertaken by a firm that are intended to result in an
innovation for the firm’.[9]
The Oslo Manual highlights 8 activities that
firms can undertake in the pursuit of innovation:
- R&D
activities
- engineering,
design and other creative work activities
- marketing
and brand equity activities
- intellectual
property (IP) related activities
- employee
training activities
- software
development and database activities
- activities
relating to the acquisition or lease of tangible assets
- innovation
management activities.[10]
The Oslo Manual notes, however, that most of these
activities can also be undertaken for reasons not related to innovation and
highlights that the ‘measurement of expenditures on innovation activities other
than R&D is an ongoing challenge’.[11]
The manual recommends that national governments make use of business
innovation surveys to collect data on business innovation (see Chapter 9).
This data is collected in Australia by the ABS with Innovation
in Australian Business results published on the ABS website as discussed
further below.
Another way to measure innovation is to construct innovation
indicators, which aggregate data from a range of sources. Indicators can be
useful for ‘exploratory analysis of innovation activities, for tracking
innovation performance over time and for comparing the innovation performance
of countries, regions, and industries’.[12]
Innovation indicators include the Global Innovation Index and the ABS
Innovation Index, both discussed below.
Key data sources
SRI
budget tables
Australian Government R&D expenditure is reported each
year in the Science,
research and innovation (SRI) budget tables. This reporting extends back to
1979-80. In recent years, the SRI budget tables have been published both as an interactive
dashboard and as a downloadable Excel
spreadsheet.
The SRI budget tables use a funder‑based approach to
measuring government support for R&D, which involves ‘identifying all those
Australia Government R&D-related programs and activities whose immediate
sources of funding is a budgetary appropriation, and then measuring or estimating
their R&D content’.[13]
The SRI budget tables use the Frascati Manual framework to
report R&D spending. The investments reported include both R&D
expenditure from departmental funds and R&D support from administered funds,
as well as revenue foregone through the R&D Tax Incentive. The tables
exclude expenditures where funding was not through an appropriation but sourced
from another organisation. This means that funding sourced from the market or
from another government agency is not included, reducing the risk of double
counting.[14]
In addition to R&D expenditure, since 2018–19 the SRI
Budget Tables have reported on non‑R&D innovation expenditure. This
is reported in the ‘Other SRI’ and the ‘KT & RC’ (knowledge transfer and
research commercialisation) spreadsheet tabs.[15]
The SRI Budget Tables provide a range of information. Examples
include:
- funding
information on R&D and innovation activities at the program or activity
level (see the ‘R&D’, ‘Other SRI’, and ‘KT & RC’ spreadsheet tabs)
- consolidated
funding information for Australia’s major R&D programs or activities (those
valued at over $100 million, such as the R&D Tax Incentive, the CSIRO,
and the Research Training and Research Support programs); figures are provided
in current and inflation adjusted prices (see the ‘Programs’ spreadsheet
tab or page 4 of the dashboard)
- figures
for Australia’s GBARD and spending on R&D tax measures
- The
Australian Government’s total investment in R&D (comprised of GBARD + R&D
tax measures). To enable comparison across time, this figure is also presented
in inflation adjusted prices and as a percentage of GDP and includes a
historical dataset back to 1978–79 (see the ‘Sector’ spreadsheet tab).
ABS R&D expenditure data
The ABS R&D surveys provide data on R&D expenditure
by business, government, higher education and private non-profit sectors.
Business
expenditure on R&D (BERD) is obtained using the biennial Survey
of Research and Experimental Development (R&D), Businesses. In addition
to total funding figures, the survey provides data on:
- BERD
as a proportion of GDP and as a proportion of gross state product (GSP) for
each of the states and territories
- BERD
by location of expenditure
- BERD
by field of research and socio-economic objective
- business
human resources devoted to R&D.[16]
BERD data is also used as an input into the ABS Australian
National Accounts expenditure item ‘Private; Gross Fixed Capital Formation –
Intellectual property products – Research and development’ (described in
further detail in Table 1).
Higher
education expenditure on R&D (HERD) is obtained using the biennial Survey
of Research and Experimental Development, Higher Education Organisations.
In addition to total funding figures, the survey also provides data on:
- HERD
as a proportion of GDP and GSP
- HERD
by field of research
- the
funding split between pure basic research, strategic basic research, applied
research, and experimental development
- source
of funding
- higher
education human resources devoted to R&D.
Government
expenditure on R&D (GOVERD) is obtained using the Survey
of Research and Experimental Development, Government and Private Non-Profit
Organisations. In addition to total funding figures, the survey also
provides data on:
- GOVERD
as a proportion of GDP and GSP
- GOVERD
by location of expenditure
- GOVERD
by field of research and socio-economic objective
- funding
sourced from the Australian Government and an aggregate of funding from the
state and territory governments
- government
human resources devoted to R&D.
Private
non-profit expenditure on R&D (PNPERD) is obtained using the same
survey as the GOVERD figures referenced above. In addition to total funding
figures, the survey also provides data on:
- PNPERD
as a proportion of GDP and GSP
- PNPERD
by location of expenditure
- PNPERD
by field of research and socio-economic objective
- type
of expenditure and source of funds
- private
non-profit human resources devoted to R&D.
Gross
expenditure on R&D (GERD) is currently obtained using
both the most recent BERD data and time series modelling to produce
estimates for government, private non-profit and higher education sectors.
Information is also presented as GERD
as a proportion of GDP, using the most recent GDP
values available.
This figure is often discussed by stakeholders, with several science
and research
advocacy groups calling for Australia to adopt a target for GERD to reach 3% of
GDP.
The methodology explanatory notes for each R&D survey
listed in this section refer to the Frascati definition of R&D as quoted
above.[17]
For a more comprehensive definition, the ABS refers to the Frascati Manual and
the Australian
and New Zealand Standard Research Classification (ANZSRC), 2020. The ANZSRC
is a set of classifications that allows the comparison of R&D data with
other countries based on type of activity, fields of research and
socio-economic objectives.[18]
Originally developed in 2008 by the ABS and Statistics New Zealand
(StatsNZ), the ANZSRC was updated in 2020 following a joint review
by the ABS, the Australian Research Council, StatsNZ and the New Zealand
Ministry of Business, Innovation and Employment.
OECD
Main Science and Technology Indicators
The OECD
Main Science and Technology Indicators website provides summary reports and
access to tables which enable comparisons between countries based on different
measures and in different years. Indicators of R&D expenditure are derived
from the OECD Research
and Development Statistics Database. The OECD
Science, Technology and Innovation Scoreboard is a tool that allows users
to generate charts of interest. Specific indicators that can be selected
include:
- GBARD
as a percentage of GDP
- total
government support of R&D (GBARD plus tax incentives)
- GERD,
BERD, HERD, GOVERD and PNPERD as a percentage of GDP.
ABS innovation in Australian business data
The ABS collects innovation
in Australian business data using the Business
Characteristics Survey (BCS), which draws on the Oslo Manual. The
BCS was recently redeveloped, taking into consideration recommendations from
the 2018-19 Innovation
Metrics Review. The
BCS now comprises an Innovation Module and a Digital Activities Module, with
data to be collected in alternate years.
The ABS has described efforts to develop a composite
indicator, the Innovation
Index, using data collected in the BCS Innovation Module. Ratings for
3 separate innovation factors, ‘novelty of goods and services (products)
or processes developed, management capability regarding innovation and
collaboration activity related to innovation’, are combined to form a single
Innovation Index measure. Each of the separate factors and the Innovation Index
are given a rating of very high, high, medium or low.
Australian
Innovation Statistics
In October 2024, the Department of Industry, Science and
Resources published the Australian
Innovation Statistics (AIS). This webpage provides innovation statistics
relating to: the domestic business environment; research collaboration; skills;
international trade; R&D expenditure; and innovation outcomes. The AIS also
has 2 ‘innovation insights’ articles providing analysis of the current state of
the Australian innovation system. The AIS appears to replace the Australian
Innovation System Monitor (and the preceding Australian
Innovation System Reports) which published annual innovation statistics
between 2010
and 2021.
Global Innovation Index
Since 2012, the World
Intellectual Property Organization has published the annual Global Innovation
Index (GII). The 2024 edition of the GII includes information on the
innovation system of 133 countries. Each country is provided with an
overall score and ranking, as well as a ranking in each of 2 sub-indexes,
7 pillars, and 78 indicators.[19]
Each country’s overall GII score is based on its average for
the following 2 innovation sub‑indexes:
- The
innovation input sub-index, which measures ‘elements of the economy that enable
and facilitate innovative activities’.[20]
The input sub-index includes indicators measuring performance across the
pillars of institutions, human capital and research, infrastructure, market
sophistication, and business sophistication.[21]
- The
innovation output sub-index, which measures ‘the result of innovative
activities within the economy’.[22]
The output sub-index includes indicators measuring performance across the
pillars of knowledge and technology outputs, and creative outputs.[23]
In addition to the annual global report, economy briefs
for Australia and other countries included in the analysis can be viewed and downloaded
from the GII Innovation
Ecosystems & Data Explorer 2024 webpage.
Table 1 Summary
of R&D metrics
| Metric (measurement
method) |
Definition (quotes from Frascati Manual unless otherwise noted) |
Source |
| BERD Business enterprise expenditure on R&D (performer‑based
approach) |
BERD ‘represents the
component of GERD incurred by units belonging to the Business enterprise
sector’ (p. 365). The business enterprise
sector comprises both private and public enterprises including:
- all
resident corporations, regardless of the residence of their shareholders, and
all other types of enterprises (not only legally incorporated enterprises)
set up to engage in market production for the financial benefit of their
owners
- ‘unincorporated
branches of non-resident enterprises are deemed to be resident because they
are engaged in production on the economic territory for a long-term basis’
- non-profit
resident enterprises offering goods or services to market. (p. 366)
|
BERD is calculated by the ABS. BERD is reported to the
OECD via the Research and Development Statistics and is available for international comparisons at
the OECD’s Main Science and Technology Indicators (MSTI). |
GBARD Government budget
allocations for R&D
(funder‑based approach) |
GBARD encompasses ‘all
spending allocations met from sources of government revenue foreseen within
the budget, such as taxation’ (p. 370). GBARD does not include
R&D financing by public corporations using funding raised in the market
(p. 370). Also, it does not include tax relief to corporations for
R&D. The SRI Budget Tables note that although
Frascati states that GBARD is intended to include all levels of government,
the SRI Budget Tables do not include state and local government funding as
that data is not available (see ‘About SRI’ spreadsheet tab). As GBARD is a funder‑based approach, it differs
from performer‑based approaches
such as GERD and GOVERD. Frascati outlines the differences between these
metrics on pages 340–1. |
GBARD is reported in the SRI Budget Tables. GBARD is reported to the
OECD via the Research and Development Statistics and is available for international comparisons at
the OECD’s MSTI.
Also included are international comparisons of total government support
(GBARD plus tax credits). |
| GERD Gross expenditure on R&D (performer‑based approach) |
GERD is the ‘total
intramural expenditure on R&D performed in the national territory during
a specific reference period’ and is the main statistic used to describe a
country’s total R&D expenditure (p. 111). GERD is an estimated figure
based on calculated BERD, GOVERD, HERD and PNPERD values. GERD can also
include ‘Rest of the world’ expenditure, which refers to domestically
performed R&D funded from abroad (although this does not appear to be
included in ABS data). GERD does not include domestic funding for R&D
performed overseas (pp. 111, 298). In addition to being
reported as a total dollar figure, GERD is reported as a percentage of GDP to
assist in international comparison. |
GERD is calculated for
Australia by the ABS. GERD is reported as part of the ABS BERD release. GERD is reported to the
OECD via the Research and Development Statistics and is available for international comparisons at
the OECD’s MSTI. |
| GOVERD Government expenditure on R&D (performer‑based
approach) |
GOVERD ‘represents the
component of GERD incurred by units belonging to the Government sector’
(p. 371). GOVERD includes federal,
state, and local government sectors and non-profit institutions controlled by
government units (p. 371). It does not include expenditure by higher
education providers (HERD) or by government business enterprises (which would
be included in BERD or in some cases HERD – see pp. 237–40). Also,
GOVERD may not include work undertaken by consultants on behalf of government
(p. 248). GOVERD should not be
confused with:
- expenditure
by the ‘public sector’ or ‘government financed GERD’ (see p. 252), which
would include both public higher education providers and government business
enterprise sector
- GBARD
(see entry above), which is a funder‑based approach.
|
GOVERD is calculated for
Australia by the ABS. GOVERD is reported to the
OECD via the Research and Development Statistics and is available for international comparisons at
the OECD’s MSTI. |
| HERD Higher education expenditure on R&D (performer‑based approach) |
HERD ‘represents the
component of GERD incurred by units belonging to the Higher education sector’
(p. 372). This includes R&D
spending by all enterprises that engage in tertiary education as well as
research institutes that are under the control or administration of tertiary
education providers. There are multiple complexities in identifying the
proportion of university income devoted to R&D as opposed to education or
administration, as discussed in Chapter 9 of Frascati. |
HERD is calculated for
Australia by the ABS. HERD is reported to OECD
via the Research and Development Statistics and is available for international comparisons at
the OECD’s MSTI. |
| Intramural R&D expenditures |
Intramural expenditures
includes all ‘current expenditures plus gross fixed capital expenditures
for R&D performed within a statistical unit during a specific period of
time’ (p. 374). Examples of statistical
units could include businesses, higher education providers, or nations.
Intramural expenditure includes all expenditure undertaken by the unit
regardless of the source of funds; this differs from ‘internal expenditure’,
which would not include R&D activities funded from an external source.
The summing of intramural spending by units within a sector produces sector‑based
statistics (such as BERD or HERD), and the summation for all sectors produces
GERD statistics. |
|
| PNPERD Private and non-profit expenditure on R&D (performer‑based
approach) |
PNPERD ‘represents the
component of GERD incurred by units belonging to the Private non-profit
sector’ (p. 377). This sector is comprised
of ‘non-profit institutions serving households’. Examples can include
charities, professional and learned societies, and non-government
organisations. This sector is residual in nature, in that it includes
non-profit organisations that cannot be included in HERD or BERD figures. See
Chapter 10 of Frascati, particularly Table 10.1 (p. 289), for
further information on the types of organisations included in PNPERD figures. |
PNPERD is calculated for
Australia by the ABS. PNPERD is reported to
OECD via the Research and Development Statistics and is available for international comparisons at
the OECD’s MSTI. |
| Private gross fixed capital formation (intellectual property products –
research and development) This is an expenditure
item reported in the Australian National Accounts |
The ABS Concepts, sources and methods publication provides a comprehensive description of
gross fixed capital formation of R&D, and explains how annual private
gross fixed capital formation – intellectual property products amounts are
determined. BERD data is identified as one of the inputs used in R&D
expenditure estimate calculations (pp. 297–298). The ABS Australian National Accounts: National Income,
Expenditure and Product publication includes a time series spreadsheet, ‘Table 3. Expenditure on Gross Domestic Product
(GDP), Current prices’, that
contains quarterly ‘Private; Gross Fixed Capital Formation - Intellectual
property products - Research and development’ figures dating back to 1959
(see ‘Data1’ spreadsheet tab, column BJ). A similar time series
spreadsheet, ‘Table 2. Expenditure on Gross Domestic Product
(GDP)’, contains annual ‘Private,
Gross Fixed Capital Formation - Intellectual property products – Research and
development’ figures dating back to 1986 (see ‘Data1’ spreadsheet tab,
column U). |
Private gross fixed capital
formation (intellectual property products – research and development) is
calculated by the ABS. Data on this metric is
published in both the annual and
quarterly
National Accounts. |