Hon Dr Michael Cullen
Deputy Prime Minister
Minister of Finance
Minister of Revenue
Leader of the House
SPEECH NOTES
Speech to Property Council of New Zealand
Thank you for your invitation to speak to you today on possible changes to the rules on property depreciation for tax purposes.
I am sure by now many of you will all have read the recent issues paper on the topic, "Repairs and maintenance to the tax depreciation rules".
While these are early days yet in the consultation process, there has been some public comment on the issues paper which I feel merits comment.
One of the more common responses has been that it is good the government is favouring investment in 'productive' activity such as electronics or machinery over 'non-productive' investment in buildings and other structures.
I personally doubt whether it makes sense to make too strong a distinction between productive and unproductive assets. When a firm purchases a machine to make shoes and this is seen to be productive, it still needs someone to build a roof and walls to provide shelter for the machine and its operators.
Those who argue that commercial and industrial buildings are unproductive have probably never tried producing goods and services in the rain, unless they are farmers, of course!
A better way to think about things, and the approach taken by the issues paper, is to ensure that our tax system neither encourages nor discourages particular types of investment, but instead allows investment to flow to where it gives the highest pre-tax return.
The underlying idea is that in the absence of taxes, there would be incentives for investment to take place in the most profitable areas. It is undesirable for taxes to drive investment decisions. This approach should provide confidence to those investing because it provides more certain ground on which to make investment decisions.
Another common response has been to refer to the complexity of the issues paper. I can respond that one of the problems with tax is simply the incredible complexity of the issues that can arise when you look at them in detail – particularly when you start with a first principles approach, as my tax policy officials are wont to do.
The issues paper on depreciation was prepared by tax policy officials, rather than the government, so does take a very high-level, first principles approach. This has led to some mild criticism as to the proposals not having much detail. The reason is that the document does not make detailed proposals but simply suggests possible ways of reducing possible tax biases in the rules and resolving some of the practical problems with their application.
The paper is a first step in the policy-making process. Once submissions have been received, officials will report to me with their detailed recommendations, and at that point the government will consider how it will consult further to ensure the detailed design of any change is also open to scrutiny.
The goal of the writers of the issues paper is modest – they want to avoid having depreciation rates which are too badly wrong. We know is not possible to get tax depreciation rates absolutely right. We know that assets do not decline predictably in nice smooth curves till the last day of their economic lives, at which point they helpfully implode into a neat pile. But we have identified some areas where the rules may be able to be improved.
Officials have argued that the starting point for neutral depreciation provisions is getting depreciation rates as close as possible to measuring how assets would depreciate in the absence of any inflation. The problem is getting as close as possible to this outcome when there is lack of data on how assets actually depreciate.
Take the case of buildings, a topic which I suspect is dear to your heart. At present, depreciation provisions are set assuming that buildings depreciate smoothly to 13.5 per cent of their acquisition cost over their economic life. This converts to a 4 per cent diminishing value rate of depreciation for most buildings. Is this too high or too low?
Some would argue it is much too high. Most properties appreciate rather than depreciate, and of course this is a key reason why many choose to invest in property.
The fact that properties rise in value would not be grounds for denying depreciation deductions on property if appreciation were due to rises in land values, with the value of structures falling over time. But it is difficult to identify changes in value that are attributable to land and those that are attributable to buildings.
If one looks at the specific example of residential accommodation and examines rateable valuations, it appears that in recent years not only has land risen in value, but the average value of structures has increased. Moreover, this increase has been considerably faster than the rate of inflation as measured by the Consumers Price Index.
Looking at these figures, one might question whether buildings would fall in value in the absence of inflation. On the other hand, it is clear that, on occasion, buildings are destroyed to be replaced with newer buildings. Clearly, by this time, the buildings have fully depreciated.
One problem with rateable valuations data is that, over time, fancier new houses or additions and extensions to existing houses boost the average value of buildings.
In the absence of good available data on New Zealand property, one approach is to examine overseas studies. The United States Treasury, in a study carried out four years ago, suggests diminishing value rates for buildings in the range of 2 to 4 per cent.
Allowing buildings to be written off in a straight-line fashion over their economic life (currently estimated to be 50 years) would lead to straight-line depreciation of 2 per cent a year. The diminishing value equivalent would be 3 per cent a year.
Of course, there are always problems in relying on overseas studies. Officials are currently examining whether it is possible to examine rateable valuations data on buildings which have not had approvals for structural improvements – to test whether a 3 per cent rate would be the best possible rate.
The paper also considers the effects of inflation. In my view, the most surprising issue in the paper is the impact of even small rates of inflation on incentives to invest. With 2 per cent inflation and a 5 per cent cost of borrowing, an investment that depreciates at 50 per cent a year has to make a 4.95 per cent return to be a break-even investment. An asset that depreciates at 2 per cent a year needs to earn only a 4.16 per cent return.
Explaining this fact, and reducing the impact of inflation on investment decisions, drives much of the analysis in the paper. The outcome is interesting in that the 20 per cent economic loading, previously seen as a concession, is now seen as a tool in addressing inflation.
The issues paper suggests increasing the loading for shorter-lived assets and reducing it for longer-lived assets, to equalise the returns required from assets that have different economic lives.
The paper looks at a number of specific issues relating to the operation of the current depreciation rules. They include the powers of the Inland Revenue to make special tax depreciation rate rules, and the tax treatment of losses on disposal of buildings and other structures.
On this last point, the paper suggests allowing a loss on destruction of buildings when the destruction is 'involuntary' – say, as the result of a natural disaster such as a flood. It also sets out the arguments for and against allowing a loss on destruction of buildings when the destruction is voluntary. In principle, there is scope to do this, though there are concerns about the potential for tax abuse. Your organisation may have some very worthwhile views on the matter.
The paper examines the growing practice on the part of property owners to break buildings into sub-components for depreciation purposes. The focus is on rental housing, rather than commercial property, since the internal layout of a commercial building is often tailored to the specific requirements of the occupant.
Increasing numbers of owners are apparently separating out from the building components like electrical wiring, internal walls and plumbing. They are then depreciating these components at specified rates, which are higher, and using the building depreciation rate for the remaining shell.
This boosts the average depreciation rate. In principle, the more assets that are broken out and depreciated separately, the lower the depreciation rate should be on the remaining shell.
The issues paper suggests two possible ways of dealing with the practice of separating out the components of rental housing. Both include listing certain structural components – such as wiring, ducting, plumbing and internal walls – and treating them as part of the building. This would limit owners' abilities to boost their buildings' depreciation rates.
The first option is similar to Australian measures for rental housing. It would allow separate depreciation of non-core chattels and fixtures when they are separately identifiable assets and are listed and separately accounted for.
It would allow equipment such as hot water cylinders and air conditioning units to be depreciated separately, as well as items that are not part of the internal structure and can readily be replaced – such as carpets, curtains and light fittings.
A cost to taxpayers of this option is that they would need to have private valuations of such assets when they bought or sold properties.
The second option would involve fewer compliance costs. Owners could choose to depreciate non-core assets such as hot water cylinders and carpets at the building rate. Everything – building and non-core assets – would be treated as one big, composite asset.
This would remove any need for private valuations of assets when rental properties are bought or sold. Another advantage is that replacements of small items such as hot water cylinders could reasonably be thought of as repairs to the large composite asset, with there being greater scope for deducting expenses as repairs and maintenance.
These, then, are the main points covered in the issues paper, as they affect property owners.
The paper is part of the government's Generic Tax Policy Process, which builds consultation with affected taxpayers into the key stages of policy-making – long before any changes are legislated for. The reason for doing this is to achieve better, more effective tax policy. We try to get it as right as possible in what is a very complex area.
The government is not proposing changes to the depreciation rules just yet – instead we are soliciting feed-back on suggested directions of change. When submissions on the issues paper are analysed, officials will make detailed recommendations to the government.
The decisions that emerge from the process will be important ones that will probably affect everyone here today. For this reason, it is vital that your organisation and other interested parties tell us their views on the suggestions put forward in the issues paper – whether or not they are workable, practical, achievable or even sensible.
I understand that submissions have started to roll in, well before the closing date, so there is obviously a lot of interest in the subject and in the issues paper.
I look forward to reading your considered views on the matter.
Thank you.