Malcolm Turnbull (Wentworth, Liberal Party) Share this | Hansard source

This legislation, the Tax Laws Amendment (Research and Development) Bill 2010 and the Income Tax Rates Amendment (Research and Development) Bill 2010, proposes altering the basic structure of Commonwealth government incentives for business research and development spending for the first time in almost a quarter of a century. If passed, it would replace the existing R&D tax concession available to businesses since 1986 with a new R&D tax credit, effective from 1 July 2010.

The existing tax concession has been the key mechanism supporting business innovation for 24 years under both Labor and coalition governments. It has been repeatedly refined and amended over the years, but is very familiar to businesses and currently assists around 8,000 firms. The concession is estimated to have a cost to the budget of $1.5 billion in the 2009-10 financial year.

Not only is the existing concession well understood by business; it also has been very successful over time in encouraging a higher level of private investment in innovation in Australia. So it is not surprising that the Rudd government’s proposal to replace it with an entirely new incentive scheme has been contentious, most of all among those Australian businesses most committed to investing in science and innovation.

The government is presenting its proposed shift to an R&D tax credit as a ‘more generous, more predictable, and less complex tax incentive’ that provides better value for money. I agree, as I am sure all honourable members would, that taxpayers have every right to demand value for money when they support the business sector. And there are a number of features of this legislation that are sensible, in particular the ideas of replacing tax deductions with tax credits and making tax credits for eligible R&D refundable in cash to small firms without tax liabilities. But in the end, the coalition opposes this legislation for the same reasons that many innovative Australian businesses have opposed it. Firstly, the government’s claims that these changes would be revenue-neutral over the forward estimates do not seem credible. While it is true the level of assistance for eligible spending will increase modestly, the activities eligible for support have been drastically narrowed. The bills look more like a revenue grab than a reallocation of resources. Secondly, Labor claims that the new regime will be simpler and easier to administer are already in tatters, given the business uncertainty and legal questions created by sweeping changes to the eligibility rules and tests, in particular the new distinction between ‘core’ and ‘supporting’ R&D and the ‘dominant purpose’ test. And thirdly, these changes replace a well-understood policy with a successful track record of encouraging business activity critical to our economic future with an incentive very different in design which may deliver very different outcomes. These risks posed were summarised very well by AVCAL, the venture capital industry group, in their response to the exposure draft:

… the narrowing of the eligibility criteria for the R&D tax incentive will materially affect the future of many innovative businesses in Australia. These businesses, which would have been at the forefront of the Government’s efforts to foster home-grown innovation, will now be ineligible for the very incentives which are intended to propel innovation forward in Australia.

While it claims this legislation is revenue neutral, the government concedes it dramatically alters the types of R&D eligible for public support. This can be seen if we consider the most important features of the bill. Assistance would be directed to core R&D and away from supporting R&D—to basic research, for example, but not industrial process improvements. Supporting R&D activities could still be assisted if innovation rather than commercial advantage is their dominant purpose but, as so many businesses have pointed out, virtually all R&D serves both objectives. Dominance of innovation alone is likely to be difficult to establish.

Tests for spending eligibility would be tightened. Currently, activities must be innovative or involve high levels of technical risk to receive support. The legislation would change this requirement to ‘considerable novelty’ and ‘high technical risk’. As AVCAL’s response to the exposure draft points out:

This change to the definition may unintentionally lead to the exclusion of many genuine value-adding R&D activities that should be supported and are currently eligible for support …

The form of the incentive would change. The current tax concession applies an uplift to eligible R&D spending to create a tax deduction which varies in value if the company tax rate changes. The new legislation would replace this with a tax credit which has a dollar value uncoupled from other tax rates. In the case of small firms without tax liabilities the credit would be refundable in cash.

The level of assistance would rise. Currently, firms receive a tax deduction of 125 per cent of eligible R&D spending or a premium tax deduction of 175 per cent when spending exceeds the average of the three previous years. Under the new legislation firms that turn over under $20 million would receive a 45 per cent tax credit—equivalent to 150 per cent deduction. Larger firms would receive a 40 per cent tax credit—equivalent to a 130 per cent deduction. Because of the narrower range of activities that would be supported, a larger proportion of assistance is expected to go to small firms with a turnover of under $20 million.

It is helpful to consider these proposed changes in the context of the original rationale for the R&D tax concession and the backdrop of the government’s broader role in science and innovation. The R&D concession was introduced by the Hawke government in 1986, with broad aims to encourage Australian firms to lift onshore research and development and so become more innovative and internationally competitive. This appeared to be a daunting challenge in the 1980s, when most Australian manufacturers were struggling to survive and modernise, and local firms invested far less in innovation on average than their counterparts in other OECD countries.

Private spending on science and innovation started rising gradually, in relative terms, in the late 1980s but as recently as 1998-99 gross business expenditure on R&D in Australia was still only 43 per cent of the OECD average. Happily, over the past decade that gap has narrowed quickly. In 2006-07, the most recent year for which we have international data, business R&D spending in Australia was 80 per cent of the OECD average. If the data is adjusted for industry structure, recognising Australia’s smaller than average manufacturing sector for an OECD country, business R&D exceeded the OECD average.

In 2007-08, the most recent year of the ABS data, Australian companies spent $14.4 billion on research and development—up 85 per cent from four years earlier. Most of that investment took place in the manufacturing and mining sectors, as we would expect. Growth rates for business R&D in Australia have been particularly rapid in the last four or five years, but this spurt is just the latest phase in a long uptrend which has transformed Australia from a laggard in private R&D into a normal advanced economy. Since the existing R&D tax concession was introduced 24 years ago, business spending on innovation has grown by almost seven per cent in real terms, annually—double the growth rate of the economy.

Obviously, the R&D concession has played a critical role in all of this—almost certainly in kick-starting business investment in innovation from its moribund levels of the mid-1980s; probably in driving strong subsequent growth; certainly in making R&D less costly for firms which spent on eligible activities. It is notoriously difficult to measure the precise impact of policies such as incentives for R&D since we cannot observe what would have happened in their absence. But it goes without saying that innovation—the introduction of new or significantly improved products or processes—should be, must be, a critical policy concern for government. In the long run a nation’s capacity to innovate determines how quickly its economy can grow, how quickly its living standards can increase and how well equipped it is to surmount economic, social or environmental challenges. Climate change and the health needs of an ageing population are just two examples of public policy challenges where it is clear innovation and new technologies are going to play a major role in the solutions.

Market economies obviously provide wonderful incentives for innovation in the private sector, but the public sector also has a critical role—and not only in providing well-educated workers, fair laws, stable macro-economic and tax policies, and the other institutions and infrastructure essential for a strong economy.

As the Productivity Commission has pointed out, there are two powerful rationales for direct use of public resources in science and innovation. The first is to pay for R&D that improves the functioning of government or the quality of the products and services it delivers. This includes advances in areas from defence to transport, health and education. Obviously, not all of this R&D needs to be carried out in the public sector, but much of it needs to be funded by the public sector. The second is to pay for R&D that generates ‘spillovers’—benefits that are not easily or completely captured by the innovator but boost the economy as a whole. Such spillovers often arise from basic research or ideas and processes that are readily mimicked, adapted or diffused without any reward to the inventor. Understandably, the private sector tends to underfund such spending.

The Commonwealth government currently spends more than $8 billion a year on what can be broadly defined as research and development, or around a third of gross expenditure on R&D in Australia. The bulk of this spending—almost $5 out of every $6—is channelled through research institutes, research grants, universities or public sector agencies such as the CSIRO or the Defence Science and Technology Organisation. The remaining $1 in $6 is largely delivered through the R&D tax concession, which of course is the main instrument for encouraging private spending likely to create spillover benefits for the wider economy as well as commercial value for the investing firm.

In the past three years the design of the existing R&D concession twice came under scrutiny during inquiries into innovation policy, first by the Productivity Commission, in 2007, and then by the review led by Dr Terry Cutler, in 2008. In each case changes were recommended. The Productivity Commission found the eligibility criteria for the basic 125 per cent tax concession did not fully screen out R&D which would have happened anyway and suggested economic pay-offs could be increased by restricting it to small firms. It recommended the 175 per cent premium rate be retained for all firms but the criteria for eligible R&D spending be tightened. Finally, the Productivity Commission acknowledged that restricting the concession would increase administrative and compliance costs. The Cutler review urged that the existing R&D tax concession be converted to a tax credit of 40 per cent for larger firms and a refundable tax credit of 50 per cent for firms with turnover under $50 million. It also recommended that all R&D undertaken in Australia which met the relevant definitions be eligible for the tax credit, including R&D by foreign companies.

There is merit in both sets of recommendations, and each review raised legitimate questions over aspects of the design of the existing R&D tax concession—in particular the provision of support to some R&D which provides sufficient returns to the companies that undertake it and which would happen in any case.

The government’s response, however, combines the most onerous features of both and ignores the generous aspects of each. Cutler’s proposed 50 per cent tax credit for small firms is reduced to 45 per cent and limited to a turnover of $20 million, not $50 million. The idea of premium support for firms which invest consistently in R&D over multiple years does not make the cut. The arbitrary distinction between ‘core’ and ‘supporting’ R&D is a heavy-handed response to the challenge of screening out spending on innovation that does not need assistance, that does not pass the additionality test. And there is little doubt the proposed legislation would add to uncertainty and complexity in the near term, rather than reduce it. There are also several positive features of the legislation, as I mentioned earlier: specifically the replacement of tax deductions with tax credits, the scope for cash refunds of credits to small firms without tax liabilities, and the higher rate of assistance provided to small firms. But on balance, however, the negatives outweigh the positives.

As I noted earlier, incentives to private firms are only a fraction of the government’s total commitment to R&D—$1.5 billion out of more than $8 billion. So while it is important to get the mechanism for distributing the $1 in $6 right, it is even more important to ensure we also spend the other $5 in the most economically beneficial way. This is a much broader topic than the scope of this legislation but it is important to address nonetheless. I have always believed Australia should develop more of its own intellectual property, and a large part of my experience before politics involved starting innovative businesses which created jobs and indeed exported technology. Over the past couple of decades Australia has evolved into a much more conducive environment for innovation than it has been in the past. I have described the growth in business investment in R&D over the past two decades, which in turn has increased the cost of the concession. We can be pretty positive about the existing environment Australia presents for innovation, particularly innovation in early-stage companies. In fact, according to a survey by Ernst & Young in 2009-10, Australia ranked as the fourth most attractive market for venture capital investment, behind the United States, the United Kingdom and Canada. But there is always room for improvement.

Our combined government and private expenditure on R&D now exceeds two per cent of GDP and is above the OECD average but it still lags behind the high levels of spending on innovation in leading countries such as the United States, Japan and Israel. Many countries have tried to emulate the dynamic level of innovation seen in the US economy, but most have failed: the web of economic, educational and cultural forces and institutions that create a Silicon Valley are hard to untangle or replicate. Israel is an interesting contrarian example, however, and over the past decade has emerged as an economy increasingly driven by innovation and technology. There are more Israeli than European companies listed on the Nasdaq, the world’s main stock exchange for tech firms, and the per capita venture capital investment in Israel is twice the per capita level in the United States and 30 times the level in the EU.

As with Silicon Valley, there are many forces behind this story, some of them difficult to replicate. Favourable immigration policies and a flow of technically skilled migrants from the former Eastern Bloc, the use of military procurement to develop targeted areas of technical expertise, higher education funding that promotes interdisciplinary collaboration and university-business linkages, and high but targeted public spending on R&D all contributed. Australia shares some of these advantages and lacks others. But perhaps the most important lesson to take away from Israel’s experience is that the public sector’s role in leading and encouraging this economic transformation was not driven by direct support for private R&D. It involved effective policy design and coherent objectives across the whole of the public sector’s support for innovation—or, in Australian terms, the $5 in $6 as well as the $1 in $6.

It would be easier to accept the Rudd government’s bona fides on innovation if it were committed to better economic returns and taxpayer value for money across the whole of public spending on science and innovation. Instead, we find it focused on narrowing the R&D tax concession, the one part of that spending which unequivocally benefits business. So much as there are some positive features of this legislation, they are outweighed by its negatives and the very strong underlying hint of Labor antibusiness ideology. I will not be voting for this legislation.

See this speech in context