Australian manufacturing: today and tomorrow

Introduction

As Australian manufacturers move up the value chain to producing ever more sophisticated products, innovation becomes increasingly important. An important source of competitive advantage is having a new product not made elsewhere, or a better product than others produce.¹

An important source of innovation is research and development (R&D) conducted within companies. It is far from the only source—much innovation involves commercialising research by universities or ‘on the job’ improvements to processes. This chapter focuses on the R&D part of innovation not because it is necessarily more important than other aspects, but because of its policy issues.

R&D is defined by the Australian Bureau of Statistics (ABS) as:

Systematic investigation or experimentation involving innovation or technical risk, the outcome of which is new knowledge, with or without a specific practical application, or new or improved products, processes, materials, devices or services. R&D activity extends to modifications to existing products/processes. R&D activity ceases and pre-production begins when work is no longer experimental.²

There have recently been some other relevant studies on the topic of innovation and R&D. The Productivity Commission released a major research report, Public Support for Science and Innovation, in March 2007, when the committee’s hearings had almost concluded. While its analysis could not be tested at hearings, some references are made to it in this chapter. The Standing Committee on Science and Innovation has released two related reports, on Riding the Innovation Wave: The Case for Increasing Business Investment in R&D (June 2003) and Pathways to Technological Innovation (June 2006), and reference is also made to these where relevant.

These studies have provided useful information about innovation processes in manufacturing. However, the committee still heard calls for more research to be done in this area.  Professor Mark Dodgson, director of the Technology and Innovation Management Centre, University of Queensland—appearing before the committee in a private capacity, noted:

There is really a paucity of good research into this whole subject ... We have a lot of interested parties doing research that helps their case but we do not have any independent research. I would contrast that very sharply with the US, the UK and European countries, which do a lot of research.³

The importance of being innovative

Technology has long been regarded as an important aspect of economic performance. It is often regarded as a ‘third factor’ complementing labour and capital in producing output. Since the late 1980s increasing emphasis has been placed on the role of innovation and R&D in improving technology. This ‘endogenous growth theory’ has gained increasing importance in mainstream economics.⁴

Economic studies suggest that R&D can generate high returns. A review of the literature by a leading ANU academic, Steve Dowrick, led him to conclude that rates of return on R&D are over 50 per cent.⁵ The Productivity Commission, while emphasising the uncertainties, suggest they could be around 30 to 100 per cent.⁶ The OECD also concludes that R&D is an important driver of economic growth.⁷ A study found a link between measures of innovation (or successful R&D) and productivity.⁸

Witnesses also stressed the role of innovation in having manufacturers that can compete in global markets. Professor Dodgson argued:

You compete with China by doing things that they cannot, and that means being hyperinnovative—producing really exciting products that do new things, delight customers and combine services into products in new and exciting ways.⁹

International comparison of research and development

Given the apparent importance of R&D in economic performance, concerns are sometimes expressed that Australia, and Australian companies in particular, do less R&D than international competitors.¹⁰

Australian businesses spent $8.4 billion on R&D in 2004–05, of which manufacturing accounted for $3.3 billion. Adding in the $2.6 billion spent by government, $4.3 billion by universities and $0.5 billion by other non‑profit institutions, Australia’s total expenditure on R&D was $15.8 billion (of which about $3.8 billion was spent on manufacturing R&D). Almost 120 000 person-years were devoted to R&D activity.¹¹

The $15.8 billion spent on R&D was the equivalent of 1.8 per cent of GDP, which placed Australia around the middle in a ranking of OECD economies. However, as the largest OECD economies (the United States, Japan and Germany) have above average R&D spending, Australia was below the OECD average R&D spending of 2.3 per cent of GDP.¹²

Australia does not have a ‘target’ for R&D. Nor do some high R&D economies such as Japan and the United States. However, a number of economies do have explicit targets, which aim to lift their R&D spending further above the 1.8 per cent in Australia (Table 8.1).

Table 8.1 R&D spending (% to GDP): actual and target

Sources: Productivity Commission, Public Support for Science and Innovation, March 2007, p. 563; OECD, Main Science and Technology Indicators, December 2006; OECD Science, Technology and Industry Outlook 2006. Values for Australia, China, Japan, Netherlands, Taiwan, UK and US refer to 2004 and NZ to 2003.

These targets are sometimes used to argue that Australia should set such a target.¹³ However, the targets are not always well-specified. Many European countries have adopted the EU’s target of three per cent announced in its ‘Lisbon strategy’ in 2002. Sweden and Finland, which were already above this target, are aiming for four per cent, while others have set more modest targets. Canada aims to exceed the OECD average; it would obviously be impossible for the majority of OECD countries to do this. Overall, there does not appear to be rigorous reasoning behind the choice of target levels; most countries seem to just choose as a target a level a bit above where they are now. Nor does it seem that many countries are likely to achieve their targets.¹⁴

An important caveat about using international comparisons to argue that Australia does too little R&D is made by the Productivity Commission:

Comparisons of input ratios are usually a conceptually unsound basis for assessing optimal investment in R&D. Nothing says that ‘high’ input ratios are necessarily better than ‘low’ ones, since it is possible to both under- or over-invest in R&D. For most other inputs — such as labour or capital — the usual interest is not in maximising inputs per output, but rather maximising its inverse (output per input or productivity).¹⁵

A 2005 study by Davis and Tunny, two Treasury economists, splits the OECD data into components of R&D and show that Australian businesses do similar amounts of ‘basic research’ to their international peers, less ‘applied research’ and much less ‘experimental development’.¹⁶ Australian non-businesses (government, academia etc) do a relatively large amount of applied research, so that total Australian spending on applied research is comparable to that in other OECD countries. Overall then, Australia appears to do a reasonable amount of ‘R’ but falls behind on ‘D’.

This view that Australians are better at inventing than commercialising agrees with anecdotal evidence. Australians invented the atomic absorption spectrophotometer, the black box flight recorder and the orbital engine but all were commercialised overseas.

It also accords with views expressed by experts in new technologies. For example, Dr Peter Binks, chief executive officer,  Nanotechnology Victoria commented:

I am not convinced that we need more R&D performed …. The big gap is being able to support industries in evaluating new technologies … the best role, in particular, for the federal government is around creating the infrastructure to support the uptake of those technologies.¹⁷

However, this view is challenged by the Productivity Commission:

There is evidence of widespread success in commercialisation across all sectors of the Australian economy, which belies a commonly expressed pessimistic view of Australia’s capabilities.¹⁸

Another conclusion from the Davis and Tunny study is that ‘the relationship between R&D and more direct measures of innovation does not appear to be strong or stable across countries’.¹⁹

In making international comparisons, it is important to note that total spending on R&D is also a function of a country’s industrial structure. Given Australia’s industrial structure – significant mining and rural sectors and less high-tech manufacturing – even if Australia had relatively high R&D within each individual industry, it would still have relatively low overall R&D relative to GDP.²⁰ Putting it another way, adjusting for its industrial structure Australia’s R&D spending is not a significantly smaller share of GDP than the OECD average.

Conclusions

The committee encourages companies to be innovative, realising this is often a prerequisite for lifting productivity and succeeding in global markets. Innovation may result from improving or redesigning processes as well as from formal research and development. In a free market firms should undertake the innovation and research that they believe will improve their profitability.

The commonly expressed concerns that Australia allocates a smaller proportion of its national income to research and development than do other OECD countries are overstated as they fail to take account of Australia’s industrial structure.

Government support for research and development

The benefits from R&D discussed above do not of themselves justify any government intervention to encourage it. Given the benefits that accrue to companies from R&D, in a free market businesses will undertake it, in the areas they judge will be of most benefit. With hindsight, some of the expenditure will not be productive, while some will generate huge returns.²¹ This is in the nature of research and is not undesirable.

Probably the main contribution government can make to encouraging the private sector to undertake R&D is to provide a sound economic background which gives the private sector the confidence to undertake projects with long-term returns.

The justification for government measures to encourage R&D is that it has some ‘public good’ characteristics. Some of the benefits from it may accrue (or ‘spill over’) to the rest of the economy rather than just to the firm undertaking it. Dr Peter Burn, associate director, public policy, from the Australian Industry Group, described these ‘positive externalities’:

Expenditure on research and development generates external benefits. A company generates more for society than the benefits it accrues for itself, so from a social point of view there is an impediment to the optimal amount of private spending on research and development just by leaving the market to itself.²²

The parts of R&D most likely to benefit the broader community rather than just the individual firm tend to be more at the R than the D end. The ‘truly original idea’ with wide ramifications is more likely to come from pure research than process improvements. These may be most likely to come from universities and research organisations.

Among companies, technological breakthroughs may be most likely to originate from ‘start-ups’ than from established companies.²³ For example, in the field of nanotechnology, it was noted that work was being done in the field by universities, government research organisations and small companies whereas:

Companies like BlueScope Steel, Amcor, BHP Billiton and Rio Tinto are all looking at nanotechnology activities, not necessarily investing right now but keeping a watching brief and developing relationships.²⁴

Dr John Raff, deputy chairman and founder of Starpharma Ltd:

Would like to see far more, I suppose, nurturing and love for the smaller structures, which are the innovators coming through, than for the larger, established organisations.²⁵

Forms of government support for R&D

Governments support R&D in six broad ways. Firstly, they directly fund research work by universities and institutions such as the Commonwealth Scientific and Industrial Research Organisation (CSIRO), Australian Nuclear Science and Technology Organisation and Defence Science and Technology Organisation. These tend to concentrate on research and do less development work.

It is sometimes suggested that public investment ‘crowds out’ industry R&D, for example, by drawing scientists away from industry. However, the available empirical evidence seems to suggest that public R&D is more of a complement to private R&D than a competitor.²⁶

Secondly, governments encourage R&D by allowing companies monopoly rights for limited periods over innovations arising from their R&D. This is done through creating intellectual property rights through patents and copyrights. Some submissions said that getting patents was a slow process in Australia, particularly if there was a challenge to them.

Thirdly, governments encourage businesses R&D by providing grants.²⁷ Fourthly, they provide tax concessions. Grants and tax concessions are discussed in the following two sections.

Fifthly, governments may support R&D by encouraging foreign firms to conduct some R&D in the domestic market. Australia is unlikely to become a major global R&D centre if it just relies on local companies. The Department of Industry, Tourism and Resources has noted that Australia is home to only two of the top 1 000 global corporate R&D spenders.²⁸

Attracting multinationals to conduct R&D in the domestic economy was an important element in the Irish ‘success story’ (discussed in Appendix F). Inward R&D by foreign firms is encouraged by Invest Australia. While an important consideration for multinational companies deciding where to locate R&D activities is the cost, the House of Representatives Science and Innovation Committee noted:

A second consideration for major international corporations is the extent to which a country is ‘innovation friendly’. This involves factors such as the availability of university, government and commercial research institutions, ‘the availability of a world-class telecommunications infrastructure’ and ‘the capacity for IT enablement’, the presence of ‘leading-edge customers to stimulate demand [and] innovation’, ‘world-class skills availability with a focus on excellence in vocational training/teaching/research’, ‘culture training in entrepreneurship’ and the availability of companies ‘that can function as partners when the R&D project requires the use of outside resources’.²⁹

Finally, governments may persuade firms to undertake more R&D in other ways. At a public hearing in Melbourne, NEC told the committee that, in the past, they were induced to place some R&D activity in Australia by ‘more or less a stick … [being told to] do the necessary conditions or you do not get government business’. They thought ‘probably the days are over’³⁰ when this approach could be used, in part as it would now be inconsistent with commitments in trade agreements.

Another witness thought there were still governments who intervened to encourage R&D in these sorts of ways. Professor Dodgson stated:

The presumption that the state does not intervene in manufacturing in other countries is seriously wrong. American manufacturing depends completely on American government policies; procurement is one example.³¹

Not all these forms of support involve government outlays and the costs of some are hard to measure. The Productivity Commission’s estimate is that ‘total funding of science and innovation by the Australian Government has actually fallen slightly as a share of GDP between 1981‑82 and 2005‑06’.³²

Grants for research and development

• Innovation Investment Fund programme (a venture capital programme discussed in Chapter 5);
• Commercial Ready;
• Commercialising Emerging Technologies;
• Industry Cooperative Innovation Programme; and
• Intermediary Access (a new programme announced in May 2007).

Commercial Ready (CR)

CR is a merit-based suite of grant programmes for SMEs, supporting innovation and its commercialisation. It provides matching grants to support R&D, ‘proof of concept’³³ work and early stage commercialisation. The scheme provides annual grants from $50 000 to $5 million.

CR began in 2004 as part of the Backing Australia’s Ability³⁴ strategy with a $1 billion programme allocation. In the 2007‑08 Budget it was extended until 2011 with a further $32 million programme commitment for applicants applying for funding up to $250 000. The grants are competitive, with one of the five criteria on which projects are assessed being the ‘national benefit’ of the project and another that it would not proceed ‘satisfactorily’ without the support. Successful applicants receive up to half the project cost, subject to a ceiling of $5 million.

It appears the CR programme helped Starpharma, a dendrimer nanotechnology company, to commercialise. Starpharma stated:

We have had a lot of support from the Australian Government through R&D Start programs, Commercial Ready and P3.³⁵

The Productivity Commission concluded that ‘there is robust evidence indicating that the Commercial Ready programme supports too many projects that would have proceeded without public funding assistance’ and recommended changes to the governance of the programme.³⁶ However, it noted that a number of organisations, such as the Australian Industry Group and the Department of Industry, Tourism and Resources (DITR), disputed this conclusion.

Science Industry Australia (SIA) wants CR extended to cover larger companies:

This turnover criterion which confines eligibility … to SMEs with an annual turnover of less that $50m is unrealistic for science industry companies that operate in the global business environment and rely on world class innovation for their competitive advantages … the relatively few larger Australian science industry companies that compete in world markets and contribute to Australia's economic and social welfare are denied access to many Government innovation support measures.³⁷

This agrees with the report of the Standing Committee on Science and Innovation, Pathways to Technological Innovation, June 2006, which recommended the government review the thresholds. This issue appears to have been ameliorated by an increased threshold to an annual turnover over $100 million, announced by the Minister for Industry, Tourism and Resources, Mr Ian Macfarlane, in August 2006.

A more effective use of grant funding may be possible if successful firms returned some of the profits resulting from R&D to the scheme. This could also discourage firms from applying for grants for projects they would still undertake without the grant.

One approach would be for the government to receive an equity stake in the project. This could work like the venture capital scheme in Israel whereby the government will ‘fund the R&D side of it and then they can be bought out by the private sector’.³⁸

Another, perhaps simpler, approach along similar lines would be to have some R&D grants take the form of income-contingent loans, modelled on the HECS scheme for university students.³⁹ These would be repaid if the R&D results in commercial success. This could have three advantages. Firstly, as earlier ‘grants’ are repaid, more funds would become available for new ‘grants’. Secondly, it would be more likely to lead to ‘additionality’; firms would not bother applying for loans for conservative projects with guaranteed returns they would undertake anyway, but would apply for more marginal, risky and innovative projects, which may be those more likely to generate spill-over benefits. Thirdly, it would build in automatic monitoring of whether the assistance scheme was succeeding in generating a reasonable number of commercial successes. The Productivity Commission note that ‘repayable schemes have been widely used overseas, and in some countries they are major forms of R&D support’.⁴⁰

Commercialising emerging technologies programme (COMET)

The COMET programme provides small innovative businesses with a package of assistance to enable them to commercialise their activities. The programme is delivered by paying for private sector business advisers who offer advice on linkages to other parties for collaboration and venture capital. The total financial assistance available under COMET is capped at $120 000 per customer.⁴¹

COMET started under the umbrella of Backing Australia’s Ability in 2001, with $40 million in funds to 2004‑05. In 2004 it was extended to June 2011 under Backing Australia’s Ability—Building Our Future Through Science and Innovation and allocated an additional $100 million. The DITR website states:

This means some 200 companies per annum can be supported with mentoring and commercialisation management advice; ... In the five years to July 2004, firms supported under COMET raised around $275 million in capital and over 500 strategic alliances, licenses and other agreements to enable their businesses to grow.⁴²

There was a suggestion that successful CR and/or COMET grant applicants could be linked up with the CSIRO to see if applied research activities could be undertaken. CSIRO commented:

There is no special relationship that CSIRO has in that process of obtaining grants with AusIndustry. I think it is fair to say, though, that CSIRO is looking at ways in which we can more effectively work with industry, especially SMEs, I might add, who often depend upon Commercial Ready and COMET grants and the like.⁴³

Industry cooperative innovation programme (ICIP)

The ICIP aims to assist business collaboration on product or process innovation, particularly between manufacturers. This programme was announced in 2004 and has funding until 2011 with a commitment of around $25 million.⁴⁴

Innovation projects may include project scoping through to implementing innovation in production. There must be a consortium of at least three entities and the applicant (at least one member of the consortium) must be a non-tax exempt company incorporated in Australia. The programme provides funding of up to $3 million.

The ICIP was praised by the Australian Electrical and Electronic Manufacturers Association as enabling collaborative research to be undertaken by advanced manufacturers, the results of which fed into their industry action agenda, but they had concerns that it did not assist companies wanting to collaborate overseas.⁴⁵ The programme has never actually excluded international entity collaboration but as this was not explicitly stated in the guidelines, there was a misconception that the programme funded domestic collaboration only. New programme guidelines were gazetted in March 2007 to clarify this issue.

SIA also referred to the need for collaborative international innovation:

With the internationalisation of Australian industry, and industry's greater use of open innovation, government should provide additional support to encouraging international linkages between manufacturers and offshore R&D.⁴⁶

Intermediary Access Programme (IAP)

The IAP was announced in the Industry Statement 2007. The five year $20 million IAP will fund 50 per cent, up to a $50 000 cap, of the cost of intermediary services to link up to 150 SMEs with possible technology partners using ‘trusted third parties’ to protect commercial-in-confidence information. This is in line with a recommendation of the National Manufacturing Forum (NMF) for the development of ‘innovation intermediaries’. The new programme will use two providers; the InnovationXchange (a not-for-profit global knowledge network) and the Australian Institute for Commercialisation’s Techfast.⁴⁷

The South Australian Government submission supported the use of TechFast.⁴⁸ The NMF praised the work of the InnovationXchange:

It is bringing companies together to share their IP in a way that does not prejudice the privacy of their intellectual property. A manufacturing advisory service could give information about the role of that body and again help that organisation’s extension to other states which are not currently engaged.⁴⁹

It is difficult at this stage to determine whether the IAP facilitates international collaborative activities, even though the intermediary organisations sponsored by the programme operate internationally.

Industry-specific grant programmes

There are also a number of industry-specific grant programmes. Examples include the Pharmaceuticals Partnership Programme and parts of the Automotive Competitiveness and Investment Scheme (ACIS). Under the ACIS scheme, rather than a direct payment, car manufacturers can receive a rebate on the duties they pay on imported components. The rebate is equivalent to 45 per cent of eligible R&D (up to a maximum of five per cent of the previous year’s sales). A looser definition of eligible R&D is used than for other schemes. Car manufacturers are allowed to include ‘re‑engineering and modification of existing products and processes’ and this forms the bulk of the R&D supported.⁵¹ The R&D component of ACIS cost $128 million in 2004–05.⁵²

This concerns the Productivity Commission, who warn that the effectiveness of government assistance for R&D is reduced as ‘a few relatively declining sectors – such as the auto industry – have benefited disproportionately through special sectorally-specific R&D programs’.⁵³

Conclusions

The committee accepts that fundamental research which benefits the broader economy, rather than just the company undertaking it, may be undersupplied in the free market so there is a case for government support. This can be provided through a competitive grants scheme along the lines of Commercial Ready. The scheme should focus on R&D with wide benefits that would not be undertaken otherwise. It need not be limited to smaller enterprises, although they may often better fit these criteria. Consideration should be given to making contingent loans as well as grants, as this will replenish the available funds and so allow more encouragement for R&D. The scheme should be simple for firms to access, with straightforward compliance requirements.

The committee identified scope in bringing together applied research activities with small innovative manufacturers by linking the successful grant applicants of either COMET or Commercial Ready to a relevant CSIRO research area or flagship. The process could be administratively simple but provide much capacity for interaction between industry and research institutions early in product or process development cycle.

One of the biggest advantages of the ICIP programme is its international collaborative approach which, based on evidence received, has until recently been viewed by industry as domestically oriented. However, although providing funding for international efforts, the programme may not fully achieve its aims because Australian companies have little way of linking with international consortia without government facilitation.

The newly announced Intermediary Access Programme may enable better facilitative links for manufacturers entering an ICIP arrangement; but at this stage it is unclear whether the IAP provides funding for international facilitative links. At this stage the two programmes appear divorced.

Research and development tax concession

There are three elements of the current R&D Tax Concession:

• Companies can deduct up to 125 per cent of qualifying expenditure incurred on R&D activities when lodging their corporate tax return.
• Companies can deduct 175 per cent of that part of R&D exceeding its average level over the past three years under the ‘Incremental (Premium) Tax Concession’.  It was announced on 1 May 2007 that this concession would be extended to foreign-owned companies.
• Under the ‘R&D Tax Offset’ programme, small firms can obtain the full benefit of their tax concession claim, regardless of whether they are in profit or not; i.e. if they are not liable for tax, they will receive a payment.⁵⁴ This is restricted to firms with R&D between $20 000 and $1 million and with group turnover below $5 million.

The R&D tax concession was originally set at 150 per cent when it was introduced in 1985 but cut to 125 per cent in 1996. Along with the reduction to the company tax rate, this has reduced its value from 23 per cent (i.e. 0.50 of 46 per cent) of R&D spending to 7.5 per cent (i.e. 0.25 of 30 per cent) now.

Unlike grants, tax concessions apply to all R&D, regardless of its quality. Views differ about whether this is a good or bad thing. Those most sceptical about the ability of governments or their advisers to ‘pick winners’, or judge which R&D is ‘high quality’, laud supporting that R&D which companies themselves see as most beneficial. They characterise the tax concession as ‘market driven’.

Alternatively, others view such tax concessions as ‘blunt measures with no quality control’⁵⁵ and argue that firms are most likely to choose R&D that is of specific benefit to themselves rather than to the broader economy. They also warn that some of any apparent increase in R&D following the introduction of tax concessions may reflect accountants (mis)classifying more expenditure as R&D, rather than a true increase in research activity. They advocate requiring firms to compete for more targeted funding of R&D likely to have wider benefits.

Views differ about whether these tax concessions have in practice actually lead to more R&D (known as ‘additionality’) or just provided a windfall gain for companies based on R&D they would have undertaken without the concession. For example, NEC told the committee that:

Our business plans—which we submit to the board—do not include any allowance for the R&D tax concession. We do not include it because we are not confident that the Government will retain it at any particular time. It has been varied significantly over the time that we have been involved in R&D and exports. We would want a commitment covering 10 years during which its conditions would not be decreased or touched if we were to include it in our board financial documents. At the moment, it is an after-the-act collection by the accountants and it goes into general revenue. Essentially, it does not affect the R&D activity.⁵⁶

Similarly, SIA pointed out that:

Larger companies in the Australian science industry argue that the compliance costs of obtaining R&D support under the Tax Concession Scheme exceed the financial benefits it provides. With the recent reductions in company tax, any benefits have been eroded further. As a consequence, Australia's R&D support measures have little impact on the competitiveness of the larger Australian-based science companies.⁵⁷

During site visits, the committee heard from some companies that tax concessions do not drive decisions and that grants are more effective. Other companies may hold similar views but not wish to say so, reasoning that the concession may be just removed rather than improved or replaced, and so they will face higher tax bills.

The Productivity Commission concludes in its 2007 report:

The extent to which the basic R&D tax concession stimulates additional R&D is low, particularly for large firms.⁵⁸

The Productivity Commission’s quantitative cost-benefit study of the 125 per cent tax concession is inconclusive; with a range from a net social benefit of around $230 million to a net social loss of a similar magnitude.⁵⁹

Other inquiries have heard similar views. Former Australian Chief Scientist, Dr Robin Batterham, said ‘my own opinion … is that, in the large company areas, the taxation concession is somewhat marginal in terms of any additionality of R&D’. Similarly, Productivity Commission Chairman, Mr Gary Banks, stated:

By and large the evidence seemed to be that most firms regarded the tax concession as something which gives them a little bit more of cash flow but did not really fundamentally affect their R&D decision-making. That was at 150 per cent.⁶⁰

The 1997 Mortimer Report⁶¹ recommended that the tax concession be cut to 100 per cent. It argued this was still a concession, as R&D was a form of capital expenditure which would otherwise not be able to be fully deducted in the year the expenditure was incurred (but like the purchase of a machine or building would be depreciated over the years).

It is possible that the current tax concession (or even the previous 150 per cent concession) is too small to have an effect but a much larger concession would have a significant impact. Of course, a much larger concession would imply a much larger cost to revenue, so this would represent a considerable gamble.

It has been observed that R&D expenditure tends to rise and fall with industry profitability.⁶² For example, in Australia recently, the mining industry has been very profitable and has strongly expanded its R&D.⁶³ One interpretation of this is that when profits are high, firms think ‘might as well stick it in R&D’, implying that ‘a lot of the R&D tax concession is going into industries that would have done R&D anyway’.⁶⁴

Compared to its OECD peers, Australia provides relatively generous tax concessions for R&D (but relatively little direct support).⁶⁵ A number of countries have R&D tax allowances, with the United Kingdom having a similar scheme with a rate of 150 per cent.⁶⁶ But other countries having a tax concession is no more an argument for such a concession here than noting other countries have tariffs or subsidies is an argument for having them here. Furthermore, some of these countries are questioning the usefulness of the concessions. Ireland discontinued its R&D tax allowance in 2001. Furthermore, international comparisons do not suggest that businesses in countries with more generous concessions do more R&D.⁶⁷

Expenditure on training workers also has potential spillover benefits, most obviously if the workers leave for another company after the training. Some capital expenditure on new equipment embodying technological advances may be a way in which new R&D contributes to greater productivity. Similar arguments could be made for spending to reduce carbon emissions, improve workplace safety or provide more generous parenting leave. It is not obvious that the spillover benefits from R&D are so much higher than those from other meritorious forms of company expenditure that R&D deserves to be singled out for favourable tax consideration.

Furthermore, Australia’s dividend imputation scheme will often offset the tax concession for R&D from the Australian shareholders’ viewpoint as the:

Imputation system … credits individual shareholders with the amount of company tax paid. When no company tax is paid because of the tax concession given at the company level, shareholders receive unfranked dividends and pay the tax themselves. There is a ‘clawing back’ of the tax concession, so that overall … investors looking at a company doing R&D have no incentive to provide capital to that company in addition to the incentive that it would have if it were not undertaking the R&D.⁶⁸

The incremental tax concession may generate more ‘additionality’ than the basic concession, but it may also have perverse incentives in encouraging firms to make their R&D spending more variable.⁶⁹ It does not reward consistently high R&D and firms cannot increase R&D indefinitely.

On the other hand, there are many, such as SIA, who argue the tax concessions do have an important influence:

Industry and commentators have argued that since the Government lowered the R&D tax concession from 150 per cent to 125 per cent in 1996, business expenditure on R&D as a percentage of Australia’s gross domestic product has declined.⁷⁰

Such opinions lead the ACCI to call for the concession to be increased:

Business has supported the R&D tax concession as an effective policy instrument addressing a market failure…. Business supports the restoration of the concession to 150 per cent.⁷¹

A similar view was put by the Australian Manufacturing Workers’ Union:

The axing of the 150 per cent R&D tax concession in 1996 was a major factor in manufacturing R&D, going from 10 per cent per annum real growth in the decade to the mid 1990s, to negative growth over the 1995-96 to 2001-02 period …. Restoring the 150 per cent R&D tax concession and leaving it unchanged for at least a decade warrants serious consideration.⁷²

DITR interviewed 116 firms in 2005 and concluded:

The study found that the R&D tax concession increases the size of investment in individual R&D projects, brings forward R&D expenditure on projects to enable faster completion with higher commercial results, and encourages investment in projects that otherwise would not be undertaken.⁷³

Even if the concession is effective, there are some companies which do not benefit. It is of limited use to newly established companies in innovative areas as they will not be making profits in their early years of operation and so not paying tax. This is a particular problem if, as argued above, it is these small nimble companies whose R&D has the most external benefit. The ‘tax offset’ component tries to address this problem by allowing the benefit to be cashed out, but this only applies to companies with R&D below $1 million, so does not apply to many new firms.⁷⁴

Another type of company that does not fully benefit from the scheme is one which holds its intellectual property overseas. In the Industry Statement 2007, it was announced that businesses which hold their intellectual property overseas would now be eligible for the 175 per cent premium concession, but they are still ineligible for the standard 125 per cent concession.⁷⁵ The Statement argues that:

Making Australia a more attractive place for world class innovation will boost investment, expand our skills base and help anchor the local arms of leading multinationals in Australia.⁷⁶

The House of Representatives Standing Committee on Science and Innovation, in its June 2006 report, Pathways to Technological Innovation, also examined the R&D tax concessions. Several submissions ‘questioned whether this was sufficient for businesses to actually increase their expenditure on R&D activities’.⁷⁷ They also noted concerns about compliance costs. However, the Committee also received submissions from companies benefiting from the tax concession and did not recommend significant changes to its operation for domestic companies.

All tax concessions complicate the tax system to some extent.⁷⁸ Provisions to avoid their abuse may increase compliance costs.⁷⁹ Generally, free market economists prefer tax systems with broad bases and low rates. Tax concessions, even for worthy goals, make this harder to achieve. The government reduced the R&D tax concession from 150 to 125 per cent as a move towards a more uniform and less distortionary tax system.

Conclusion

There are doubts about the extent to which the existing R&D tax concessions are effectively inducing additional R&D, especially given the reduction in the company tax rate. The concessions may not be the optimal form of incentive. Replacing the concessions with increased grants would allow for a more targeted approach, although grants may have disadvantages, such as administrative costs and risks of favouritism.

The committee recognises that there may be benefits to Australia from conducting R&D here, even if the intellectual property rights are held offshore. R&D and design activities, rather than production, will increasingly form the basis for Australia’s involvement in global manufacturing. The committee commends the recent decision to allow companies holding intellectual property offshore access to the 175 per cent premium concession and believes the same argument is applicable to the standard 125 per cent scheme.

Examples of abuse involving non-R&D activity being misrepresented as R&D to obtain the tax concession are given in Productivity Commission, Public Support for Science and Innovation, 2007, p. 378. Back