Black hole tax treatment of research and development expenditure
Agency Disclosure Statement
This Regulatory Impact Statement has been prepared by Inland Revenue.
It provides an analysis of options to address the problems with the current “black hole” tax treatment of certain research and development (R&D) expenditure.
Generally, business taxpayers will try to reduce their income tax liability by claiming deductions for business expenditure, wherever possible, against their assessable income. “Black hole” expenditure is business expenditure that is not immediately deductible for tax purposes, and also does not form part of the cost of a depreciable asset for tax purposes and, therefore, cannot be deducted over time as depreciation.
Black hole tax treatment of expenditure can produce economic distortions. A taxpayer may choose to invest in an area where they can deduct or depreciate their expenditure instead of investing in an area where they cannot. If investing in the area that receives black hole tax treatment would have been the most efficient choice in the absence of taxation, the taxpayer’s investment decision has been distorted by tax settings.
The preferred option would reduce these distortions, by allowing capitalised R&D expenditure to be either depreciated or deducted, depending on the particular circumstances.
Initial proposals to provide tax deductibility for capitalised R&D expenditure were consulted on via the release of a Government discussion document on 7 November 2013.
The discussion document proposed making capitalised development expenditure that creates an intangible asset with a reasonably certain useful life part of the depreciable costs of the asset. Submitters generally accepted that this was the appropriate way to treat this expenditure.
The discussion document also proposed allowing a deduction for capitalised R&D expenditure towards an unsuccessful intangible asset with a reasonably certain useful life when the asset is written off for accounting purposes. This proposal would have meant capitalised R&D expenditure towards intangible assets with uncertain useful lives would have remained non-deductible. A number of submitters were concerned that this would leave a significant category of capitalised R&D expenditure still never being deductible for tax purposes, and that this was not the appropriate treatment of expenditure on intangible assets with indefinite but finite useful lives. After consideration of this feedback, the proposals were altered to also make these costs deductible when the asset is written off for accounting purposes.
The Treasury and the Ministry of Business, Innovation and Employment were involved in the policy development of the options discussed in this RIS, and they agree with the conclusions and recommendations made.
There is some uncertainty around the estimated fiscal costs of the options, as significant assumptions were made in developing fiscal cost estimates, due to lack of source data and limited relevant additional information provided by submitters. There are no other significant constraints, caveats or uncertainties concerning the analysis undertaken.
The preferred option and the other alternative policy options will impose some additional compliance costs on businesses that wish to avail themselves of the proposed increased allowance of tax deductions for R&D expenditure. However, businesses would only incur these additional compliance costs in cases where they consider that the benefit to them of the increased allowance of deductions outweighs the costs.
None of the policy options would impair private property rights, restrict market competition, reduce the incentives for businesses to innovate and invest, or override fundamental common law principles.
Mike Nutsford
Policy Manager, Policy and Strategy
Inland Revenue
27 March 2014
STATUS QUO AND PROBLEM DEFINITION
Current tax rules
1. “Black hole” expenditure is business expenditure that is not immediately deductible for tax purposes, and also does not form part of the cost of a depreciable asset for tax purposes and, therefore, cannot be deducted over time as depreciation.
2. Under current tax rules, a person is allowed an immediate deduction for expenditure they incur on research or development up until an intangible asset is recognised for accounting purposes. Further development expenditure is capitalised.
3. Capitalised development expenditure can only be depreciated (that is, deducted over the life of an asset) for tax purposes once there is “depreciable property” under the Income Tax Act 2007 (ITA). Expenditure on intangible property may only be depreciated if the intangible property is listed in schedule 14 of the ITA, which lists items of “depreciable intangible property”. For an item of property to be listed in schedule 14, it must be intangible and have a finite useful life that can be estimated with a reasonable degree of certainty on the date of its creation or acquisition.
4. In the event that a research and development (R&D) project does not create a depreciable asset for tax purposes, the development expenditure that has been capitalised will be rendered non-deductible, either immediately or over a period of time. This includes capitalised development expenditure on assets that are completely unsuccessful, as well as intangible assets that are useful but are not listed in schedule 14.
5. Moreover, even if the project does create an asset that is listed in schedule 14, capitalised development expenditure incurred in creating the asset may still be rendered non-deductible, either immediately or over a period of time. As explained in paragraphs 6 and 7 below, this may occur because, although the expenditure has given rise to an asset that is depreciable for tax purposes, the depreciable costs of the asset have been interpreted to exclude development expenditure.
6. An interpretation statement issued by Inland Revenue takes the view that the depreciable patent costs (for a taxpayer who has lodged a patent application with a complete specification or had a patent for an invention granted) are limited to the administrative and legal fees incurred in the patent process.[1] According to Inland Revenue’s view of the law, capitalised development expenditure relating to the invention that is the subject of the patent (or patent application) is potentially neither deductible nor depreciable for tax purposes.
7. Although the interpretation statement is confined to patents, it is likely that the depreciable costs of plant variety rights would be interpreted in the same way, given that they are both types of intellectual property rights obtained by registration following an R&D process.
The problem
8. Black hole tax treatment of expenditure can produce economic distortions. A taxpayer may choose to invest in an area where they can deduct or depreciate their expenditure instead of investing in an area where they cannot. If investing in the area that receives black hole tax treatment would have been the most efficient choice in the absence of taxation, the taxpayer’s investment decision has been distorted by tax settings.
9. The scale of the problem cannot be quantified with any degree of precision, as we do not have direct information on what projects would have been undertaken in the absence of taxation. The vast majority of R&D expenditure is already immediately tax deductible. However, there is still room for improvement.
OBJECTIVES
10. The objectives against which the options are to be assessed are to:
(a) ensure economic efficiency by ensuring that, as far as possible, investment decisions are not distorted by tax considerations;
(b) provide certainty about the tax treatment of particular expenditures;
(c) minimise compliance costs for taxpayers; and
(d) ensure the coherency, consistency and integrity of the overall tax system.
11. Objective (a) is the key objective in this analysis because the aim of the review is to reduce the cases where tax rules may be discouraging R&D investments that would be undertaken in the absence of taxation. We recognise that there are likely to be trade-offs between these tax policy objectives. For example, the preferred option minimises economic distortions but will involve some compliance costs to ensure the integrity of the tax system.
12. It is also necessary to consider the Government’s Business Growth Agenda (BGA), which emphasises the importance of building innovation to help grow New Zealand’s economy. “Encouraging business innovation” is one of the seven key initiatives of the Building Innovation work stream, which recognises that enabling R&D is a key element in the innovation process.
REGULATORY IMPACT ANALYSIS
13. Several options have been considered for addressing the problem and achieving the stated objectives. These options are set out below.
Option one
14. Option one is to retain the status quo. Under the status quo, capitalised development expenditure will continue to be neither deductible nor depreciable for tax purposes.
Option two
15. Option two is to allow failed capitalised development expenditure, which the taxpayer intended would lead to an item of “depreciable intangible property”, to be depreciated over the estimated useful life of the asset the development expenditure was intended to create.
Option three
16. Option three is to allow an immediate deduction for failed capitalised development expenditure, which the taxpayer intended would lead to an item of “depreciable intangible property”, upon the intangible asset being written off for accounting purposes.
Option four (preferred option)
17. Option four is to allow a one-off tax deduction for capitalised development expenditure upon the intangible asset to which it relates being written off for accounting purposes, for taxpayers who have developed intangible assets that are not depreciable for tax purposes. This would apply irrespective of whether the asset was useful for a period or a completely unsuccessful investment.
Option five
18. Option five is to allow capitalised development expenditure that creates an intangible asset with an uncertain useful life to be depreciated over a given period of time. This would apply irrespective of whether the asset was useful for a period or a completely unsuccessful investment.
Option six
19. Option six is to:
- allow capitalised development expenditure that creates a useful intangible asset with an uncertain useful life to be depreciated over a given period of time; and
- allow an immediate deduction for capitalised development expenditure that gives rise to a completely unsuccessful intangible asset upon the asset being written off for accounting purposes.
Further proposals
20. Additionally, each of options two to six, would allow capitalised development expenditure that creates an intangible asset with a useful life that can be estimated with a reasonable degree of certainty at the time of its creation to be depreciated over that life.
21. As an integrity measure, each of options two to six would also involve the introduction of appropriate claw-back rules (outlined below).
22. In the event that an intangible asset that has been written off for accounting purposes becomes useful, it is proposed that any capitalised development expenditure previously allowed as a tax deduction would be clawed back as income. The clawed-back amount would then be able to be depreciated over the estimated useful life of the asset, if the asset is depreciable.
23. In the event that an intangible asset that has been written off for accounting purposes is sold, it is proposed that any capitalised development expenditure previously allowed as a tax deduction (or the sale proceeds, if this amount is lower) would be clawed back as income.
Impacts of options
24. The table below summarises the impacts of each of the options.
Table 1: Impacts of the options
Option |
Meets objectives? |
Impacts |
Net impact |
||||
|---|---|---|---|---|---|---|---|
Economic impact |
Fiscal impact |
Administrative impacts |
Compliance impacts |
Risks |
|||
One Status quo |
No | Potential for capitalised R&D expenditure to receive black hole tax treatment and this could discourage investments in R&D that would have been undertaken in the absence of taxation. |
No impact. |
No impact. |
No impact. |
None. | Does not address the problem or achieve any of the stated objectives, as it may lead to a sub-optimal level of investment in R&D. |
| Two | No | Would reduce the tax distortion against some R&D investments, but there would still be distortions as not all capitalised R&D expenditure would be covered. Economically neutral between successful and unsuccessful projects. |
Fiscal cost is unquantified, but would likely be lower than option 3, as the deductions for failed capitalised development expenditure would be spread over time rather than taken immediately upon write off. |
No systems implications for Inland Revenue, but there may be some minor one-off additional administrative costs, which would be met within existing baselines. |
Some additional compliance costs, but taxpayers would only incur them where they consider the benefit of the increased allowance of deductions outweighs them. Depreciation of failed capitalised expenditure means higher compliance costs than options 3 and 4. |
Potential perception that this option does not go far enough, as it would not provide tax deductibility for capitalised development expenditure on intangible assets with uncertain useful lives. |
Does not fully address the problem, and fails to achieve any of the stated objectives. Specific concerns include:
|
| Three | C | Would reduce the tax distortion against some R&D investments, but there would still be distortions as not all capitalised R&D expenditure would be covered. |
Under the preferred transitional approach, estimated aggregate fiscal costs of $5.3m over the period 2014/15 to 2017/18. |
No systems implications for Inland Revenue, but there may be some minor one-off additional administrative costs, which would be met within existing baselines. |
Some additional compliance costs, but taxpayers would only incur them where they consider the benefit of the increased allowance of deductions outweighs them. Immediate deduction for failed capitalised expenditure means lower compliance costs than options 2, 5 and 6. |
Potential perception that this option does not go far enough, as it would not provide tax deductibility for capitalised development expenditure on intangible assets with uncertain useful lives. |
Does not fully address the problem, and fails to achieve all of the stated objectives. Specific concerns include:
|
Four Preferred option |
A, B, C and D |
More effective than options 2 and 3 in reducing the tax distortion against R&D investments. Greatest expected improvement in productivity and growth. |
Under the preferred transitional approach, estimated aggregate fiscal costs of $13.1m over the period 2014/15 to 2017/18. |
No systems implications for Inland Revenue, but there may be some minor one-off additional administrative costs, which would be met within existing baselines. |
Some additional compliance costs, but taxpayers would only incur them where they consider the benefit of the increased allowance of deductions outweighs them. One-off tax deduction for capitalised expenditure on non-depreciable intangible assets means lower compliance costs than options 2, 5 and 6. |
Would place additional pressure on the definition of R&D and Inland Revenue’s ability to monitor the line between capitalised R&D expenditure and other capitalised expenditure. |
Addresses the problem and achieves all of the stated objectives. Overall, greatest improvement upon the status quo as it would reduce the tax distortion against R&D investments, provide the most coherence and certainty, and minimise increases in compliance costs. |
| Five | B | More effective than options 2 and 3 in reducing the tax distortion against R&D investments. Could provide a tax-subsidy to investment in R&D-generated intangible assets with uncertain useful lives. |
Fiscal cost is unquantified, but would likely be higher than options 2 and 3 due to the wider ambit of capitalised development expenditure that would be eligible for deductions, and lower than option 6 as the deductions for failed capitalised development expenditure on intangible assets with uncertain useful lives would be spread over time rather than taken immediately upon write off. |
No systems implications for Inland Revenue, but there may be some minor one-off additional administrative costs, which would be met within existing baselines. |
Some additional compliance costs, but taxpayers would only incur them where they consider the benefit of the increased allowance of deductions outweighs them. Depreciation of capitalised expenditure that creates an asset with an uncertain useful life means this option has the highest compliance costs. |
Would likely create pressures for assets with longer (but certain) finite lives to be characterised as assets with finite but indefinite lives. Would place additional pressure on the definition of R&D and Inland Revenue’s ability to monitor the line between capitalised R&D expenditure and other capitalised expenditure. |
Does not fully address the problem, and fails to achieve all of the stated objectives. Specific concerns include:
|
| Six | B | More effective than options 2 and 3 in reducing the tax distortion against R&D investments. Could provide a tax-subsidy to investment in R&D-generated intangible assets with uncertain useful lives. |
Fiscal cost is unquantified, but would likely be higher than options 2 and 3 due to the wider ambit of capitalised development expenditure that would be eligible for deductions, and higher than option 5 as the deductions for failed capitalised development expenditure on intangible assets with uncertain useful lives would be taken immediately upon write off rather than spread over time. |
No systems implications for Inland Revenue, but there may be some minor one-off additional administrative costs, which would be met within existing baselines. |
Some additional compliance costs, but taxpayers would only incur them where they consider the benefit of the increased allowance of deductions outweighs them. Depreciation of capitalised development expenditure that creates a useful intangible asset with an uncertain useful life means higher compliance costs than options 3 and 4. |
Would likely create pressures for assets with longer (but certain) finite lives to be characterised as assets with finite but indefinite lives. Would place additional pressure on the definition of R&D and Inland Revenue’s ability to monitor the line between capitalised R&D expenditure and other capitalised expenditure. |
Does not fully address the problem, and fails to achieve all of the stated objectives. Specific concerns include:
|
Fiscal costs
25. The fiscal cost estimates should be treated with some caution. Due to lack of source data and limited relevant additional information provided by submitters, significant assumptions were made in developing them, for example:
- the stock of capitalised R&D expenditure;
- the percentage of capitalised R&D expenditure that will be depreciated; and
- the R&D failure rate.
26. Inland Revenue has carried out sensitivity analysis around some of the assumptions and the fiscal costs do not vary materially.
Compliance costs
27. The proposed changes are taxpayer-friendly, but will impose some additional compliance costs on businesses that wish to avail themselves of the proposed increased allowance of tax deductions for R&D expenditure. These additional compliance costs are associated with:
- complying with a higher accounting standard than the new minimum requirements;[2]
- claiming a deduction for expenditure that previously would have been non-deductible; and
- application of the proposed claw-back rules for written off assets that become useful or are sold.
28. However, these additional compliance costs would only be imposed on those businesses that wish to avail themselves of the proposed increased allowance of tax deductions for R&D expenditure. Therefore, businesses would only incur these additional compliance costs in cases where they consider that the benefit to them of the increased allowance of deductions outweighs the costs. Furthermore, we consider that the proposed claw-back rules are important integrity measures which would not be expected to often require application.
Social, environmental or cultural impacts
29. There are no social, environmental or cultural impacts associated with any of the options considered above.
Net impact of all options
30. The preferred option (option four) addresses the problem by reducing the cases where tax rules could discourage R&D investments that would be undertaken in the absence of taxation. It also achieves all of the stated objectives.
31. Inland Revenue does not support options one, two, five and six because they do not fully address the problem and fail to achieve some or all of the stated objectives. We originally preferred option three (the discussion document’s proposal) but after consideration of the feedback received, and further analysis of that option, it is no longer preferred.
CONSULTATION
32. Public consultation was carried out via the release of a consultation document, Black hole R&D expenditure: a government discussion document, on 7 November 2013.
33. The proposals in the discussion document were essentially option three in the above regulatory impact analysis.
34. Twelve submissions were received in relation to the discussion document. The submissions were generally supportive of the intent of the proposals to relieve black hole R&D expenditure. However, many submitters were concerned that the initial proposals would still leave a significant category of capitalised development expenditure never being deductible for tax purposes. These submitters argued that this was not the appropriate treatment of expenditure on intangible assets with indefinite but finite useful lives. These submitters wanted the scope of the proposals widened to provide tax deductibility for – both successful and unsuccessful – capitalised development expenditure towards intangible assets that are not listed in schedule 14 of the ITA.
35. While it would be inappropriate, from an economic perspective, to allow tax deductibility for expenditure towards creating an asset that would not have been likely to have a finite life if successful, we recognise that technology tends to move at a relatively fast pace and that it is likely that R&D-generated assets will have limited lives, even if those lives are not capable of being estimated with a reasonable degree of certainty at the time of the asset’s creation. We were therefore sympathetic towards the submitters’ concern.
36. In order to respond to this concern, we considered alternative options that would eliminate black hole R&D expenditure on a prospective basis. This led us to alter the proposals, arriving at option four as our preferred option.
37. The Treasury and the Ministry of Business, Innovation and Employment have been consulted and agree with our conclusions and recommendations.
CONCLUSIONS AND RECOMMENDATIONS
38. We recommend:
- allowing capitalised development expenditure that creates an intangible asset with a useful life that can be estimated with a reasonable degree of certainty at the time of its creation to be depreciated over that life; and
- allowing a one-off tax deduction for capitalised development expenditure upon the intangible asset to which it relates being written off for accounting purposes, for taxpayers who have developed intangible assets that are not depreciable for tax purposes. This would apply irrespective of whether the asset was useful for a period or a completely unsuccessful investment; and
- introducing appropriate claw-back rules that would apply when an intangible asset that has been written off for accounting purposes becomes useful or is sold.
39. The proposals would enable all capitalised R&D expenditure to be deducted (thereby providing certainty of tax treatment) and would reduce the cases where tax rules discourage R&D investments that would be undertaken in the absence of taxation, but without potentially providing a tax-subsidy to investment in R&D-generated intangible assets with uncertain useful lives.
40. The proposed tax treatment of successful capitalised development expenditure on intangible assets with reasonably certain useful lives is consistent with the usual tax treatment of capitalised expenditure that has created a depreciable asset.
41. The proposed tax treatment of failed capitalised development expenditure is consistent with the usual tax treatment of failed capitalised expenditure. While the proposed tax treatment of capitalised development expenditure that creates useful assets with uncertain useful lives is unusual, it has the effect of restricting deductions to cases where it is clear that the expenditure is of no on-going value. For this reason, we prefer it to depreciating the expenditure over a given period of time, which will inevitably be too short in some cases (implying a tax-subsidy) and too long in others. As technology tends to move at a relatively fast pace, it is likely that R&D-generated assets will have limited useful lives, even if those lives are not capable of being estimated with a reasonable degree of certainty at the time of the asset’s creation. Therefore, the proposed treatment improves upon the status quo, as not allowing any deduction for expenditure that has created an asset with a finite useful life is inappropriate.
42. While there may be some additional compliance costs (as compared to the status quo) in order to get a deduction, taxpayers will only incur these additional costs where they consider that the benefit to them of the increased allowance of deductions outweighs the costs. The preferred option minimises these compliance costs by allowing a one-off tax deduction for capitalised development expenditure rather than requiring taxpayers to depreciate failed expenditure or successful expenditure on assets with uncertain useful lives over time.
IMPLEMENTATION
Transitional approach
43. We considered three options (set out in the table below) for transitioning to the proposed new rules. We note that most of these options are linked in some way to the date of release of the discussion document (that is, 7 November 2013). The reason why this date was chosen, as opposed to a prospective date, is that this latter alternative may have created an undesirable incentive for taxpayers to defer their R&D spending in anticipation of the proposed new rules.
Table 2: Transitional options
Option 1 |
Option 2 |
Option 3 |
|
|---|---|---|---|
R&D that creates a depreciable intangible asset |
Only capitalised R&D expenditure incurred from 7 November 2013 would be eligible for depreciation deductions. |
All capitalised R&D expenditure (whenever incurred) relating to assets created (that is, recognised for tax purposes) from 7 November 2013 would be eligible for depreciation deductions. |
In addition to allowing all capitalised R&D expenditure (whenever incurred) on new assets to be depreciated, pro-rated depreciation deductions would be allowed for capitalised R&D expenditure that relates to existing assets. |
R&D that does not create a depreciable intangible asset |
Only capitalised R&D expenditure incurred from 7 November 2013 would be eligible for the one-off tax deduction upon write off for accounting purposes of the intangible asset to which it relates. |
All capitalised R&D expenditure (whenever incurred) relating to intangible assets written off for accounting purposes from 7 November 2013 would be eligible for the one-off tax deduction. |
All capitalised R&D expenditure (whenever incurred) relating to intangible assets written off for accounting purposes from 7 November 2013 would be eligible for the one-off tax deduction. |
Analysis of options
44. Option 1 only gives deductions for new R&D expenditure, whereas options 2 and 3 would give windfall gains to those who have incurred sunk costs in developing assets. Therefore, option 1 is the most targeted of the three options, with options 2 and 3 providing increasing recognition that the status quo is a poor outcome under tax policy frameworks through providing relief from black hole expenditure on an increasingly wider ambit of historical R&D expenditure.
45. Although option 3 would allow the widest ambit of depreciable expenditure, there would be higher compliance costs associated with apportionment and integrity issues in relation to old documentation of costs. Option 1 could have slightly higher compliance costs than option 2, associated with the need to go back and attribute expenditure to pre- and post- 7 November 2013.
46. The annual fiscal cost of all three options would eventually converge. However, over the short to medium term, option 1 would be the least fiscally expensive, and option 3 would be the most fiscally expensive.
47. Options 2 and 3 offer an additional benefit in that they would reduce the bias that those who have incurred sunk costs developing an asset have towards selling the resulting asset over continuing to hold it. This bias exists because, currently, a purchaser of one of these assets can depreciate the entire purchase cost, which means that such assets are potentially more valuable to purchasers than to the person who has developed them.
Preferred transitional approach – conclusion
48. We prefer option 1, which only gives deductions for new R&D expenditure, because the fiscal cost incurred as a result of the proposed changes would be more closely aligned with the Government’s objective of increasing new business R&D.
49. Options 2 and 3 would give windfall gains to those who made an economic decision to proceed with developing an asset in the expectation that development expenditure incurred from the point of asset recognition for accounting purposes would be neither immediately deductible nor depreciable. These options are estimated to be considerably more fiscally expensive over the short to medium term, but would provide limited additional benefit in reducing the bias that those who have incurred sunk costs developing an asset have towards selling the resulting asset over continuing to hold it.
Further implementation details
50. If approved, the proposals will require changes to the Income Tax Act 2007. These changes would be included in the next available taxation bill after Budget 2014 and take effect from the 2015/16 income year.
51. When introduced to Parliament, commentary will be released explaining the amendments, and further explanation of their effect will be contained in a Tax Information Bulletin, which would be released shortly after the bill receives Royal assent.
52. The proposals would have no systems implications for Inland Revenue but may result in some additional administrative costs, such as costs associated with publications to communicate the changes. These costs are expected to be insignificant and would be met within existing baselines.
MONITORING, EVALUATION AND REVIEW
53. In general, Inland Revenue’s monitoring, evaluation and review of new legislation takes place under the Generic Tax Policy Process (GTPP). The GTPP is a multi-stage tax policy process that has been used to design tax policy in New Zealand since 1995. The final stage in the GTPP contemplates the implementation and review stage, which can involve post-implementation review of the legislation, and the identification of any remedial issues. Opportunities for external consultation are also built into this stage. In practice, any changes identified as necessary for the new legislation to have its intended effect would generally be added to the Tax Policy Work Programme, and proposals would go through the GTPP.
[1] Interpretation statement “Income tax treatment of New Zealand patents”, Tax Information Bulletin Vol 18, No 7 (August 2006), p 51.
[2] We note that, when the Financial Reporting Act 2013 comes into effect on 1 April 2014, minimum financial reporting requirements will be reduced for many businesses. The current tax provisions that allow a tax deduction for R&D expenditure, and the proposal to allow a tax deduction for taxpayers who have developed intangible assets that are not depreciable for tax purposes, are linked to particular accounting standards.