More than 170 countries now have a goods and services tax (GST) or value-added tax (VAT), and the tax accounts for approximately one fifth of total tax revenue worldwide. In most countries with a GST/VAT, it is levied on the first supply of goods, and it is assumed that the value of goods at this time is equal to the present value of all future consumption of those goods. Second-hand sales of goods between unregistered consumers are not included in the tax base.
From a tax policy perspective, this approach achieves the correct result where goods are immediately consumed, or their value depreciates over time. Real property is the common durable good that appreciates.
Over the long term, the value of real property generally increases, yielding more consumption which is not taxed. This is not problematic when it comes to commercial real property, as registered entities are entitled to input tax credits for the GST/VAT that they pay when they purchase or lease commercial property if it is used as an input to the entity’s production.
The current GST/VAT treatment of residential premises is therefore problematic as any appreciation in their value is left out of the tax base. This is well recognised in the GST/VAT literature. The significance of this issue lies in the number of residential premises and their value.
The issue has prompted my recently published journal article on how an alternative approach of including imputed rent in the GST/VAT base might work. While it has previously been acknowledged in the tax literature that imputed rent could be brought within the GST/VAT base, there has been little discussion in the past of what this would look like.
Imputed rent is the theoretical value that an owner-occupier receives for living in their home. If imputed rent were included in the GST/VAT base, the first supply of residential premises would not be taxable. Instead, a notional value would be placed on an owner-occupier’s imputed rent, and this would be subject to GST/VAT on a periodic basis.
First simulation
Building on earlier findings of the United Kingdom report of the Mirrlees Review, this research provides the first simulation showing how a homeowner’s annual GST/VAT liability could be calculated on a periodic basis in relation to the consumption of residential premises. It shows the potential for the GST/VAT base to be broadened so that the added value of consumption which results from the general appreciation in residential premises is included in the base.
For this simulation, Australian data is used. But, data from other jurisdictions could be inputted for future research. This could help to show the effect of shifting from prepaid to periodic GST/VAT.
It was assumed that residential premises appreciate over time at an annual rate of 5.52%. This was based on historic annualised trends in data published by the Australian Bureau of Statistics. A market rental rate was assumed based on the gross national rental yield published by CoreLogic. While Australia has recently experienced high inflation, like in many other countries, a 2.4 per cent rate of inflation was assumed as this was calculated to be the average inflation rate in Australia since 1992. This simulation used the mean GST-exclusive value of residential premises to calculate imputed rent. The Australian GST rate is 10%.
In year 1 of the simulation, the value of consumption was regarded as equal to the market rental rate multiplied by the GST-exclusive value of the residential premises. The GST liability was calculated to be the value of consumption multiplied by the GST rate. In later years, it was assumed that the value of residential premises would rise by the annual appreciation rate, so do the value of consumption and the GST liability. The present value of future GST payments was calculated as the GST liability for the year divided by (1+0.024)x.
Assuming no change in the values of the variables used in the simulation, comparing the accumulated present value of GST liabilities under the periodic approach to the GST payable under the prepaid approach showed that for the first 21.2 years GST collected would be the same in present value terms to the GST that would be collected if the prepaid approach were retained. More consumption would be taxed from year 22 under the periodic approach.
Table 1: Pros and cons of the prepaid and periodic approaches
Possible downward effect on price of residential premises and leases
If the periodic approach was implemented, the upfront cost of purchasing residential premises, including for both newly-built and subsequent resales, would reduce. This is because while GST/VAT is currently only paid in relation to the first supply of residential premises, it is likely to be carried forward in the price of subsequent sales and in the cost of residential rent.
Under the periodic approach, the impact of the GST/VAT liability would be spread over many years. This might have a downward effect on the price of residential premises and leases. This might contribute towards addressing housing affordability concerns, currently experienced in many jurisdictions.
Implementation issues
There are implementation issues relating to collecting GST/VAT on imputed rent. However, collection is possible.
Some homeowners might prefer to pay GST/VAT annually if they do not have the cash to pay upfront GST/VAT under the current approach. However, just like under the current approach, paying GST/VAT periodically could be difficult for those who have a low income or no income.
Options for addressing these concerns include deferring the GST/VAT liability, providing social assistance for those who are particularly disadvantaged, and introducing progressive GST/VAT rates dependent on the purchase price.
There would be increased costs to the tax administration relating to GST/VAT collection, but this could be funded through more residential premises coming within the GST/VAT base that have never been included (in the case of Australia), and through the appreciation of residential premises also being included within the base.
The politically easiest transition rule would be to only apply the alternative periodic approach to residential premises purchased after the implementation of this approach, and to leases which commence after these residential premises are purchased.
Rules can be introduced so that homeowners pay GST/VAT directly to the tax administration, without requiring them to register for GST/VAT. The tax administration could calculate the homeowner’s tax liability based on the initial price of the residential premises, informing the homeowner annually of the amount of GST/VAT due. Homeowners should not be entitled to input tax credits for GST/VAT paid in relation to their consumption of residential premises.
Homeowners should be required to register for GST/VAT when they lease residential premises to tenants. This should be the case independent of whether they would otherwise meet the registration turnover threshold. They should charge GST/VAT on residential leases and be entitled to input tax credits in relation to GST/VAT paid on their inputs, including in relation to imputed rent, repairs and maintenance.
Over the long term, a shift to the alternative periodic approach would lead to more revenue collection, but in the short term there would be a reduction in revenue collected. This shortfall could be funded through adjustments to the transitional rule. Another alternative is to implement wider tax reform that might address this temporary tax shortfall. This includes reforming the income tax system and broadening the GST/VAT base to include items such as financial services.
There would be political challenges to overcome in implementing a periodic approach to GST/VAT on the consumption of residential premises. However, there are ways to make this alternative approach more acceptable to homeowners. For example, the removal of stamp duty on transfers of residential premises or land tax (in jurisdictions that have these) may be a way.
This article is based on Peacock, Christine, (2023), Shifting from pre-paid to periodic GST on the consumption of residential premises, Australian Tax Forum 38(2), pp.199-223.
Houses do not appreciate. They depreciate and need repairs as any homeowner knows. It is land which appreciates.
So why not forget about taxing “valued added” – after all, don’t we WANT increased productivity?
Just levy a universal ad valorem tax on all land values (urban, agricultural, minerals. fisheries, spectrum etc) and give a non-refundable credit against income tax for homeowners.
By the way, a lot of housing expenditure is NOT consumption. The sheltering and reproducing of labour is to labour as a factor of production what a factory roof, grease and oil, repairs and replacements are to physical capital.
So called consumption taxes are very much taxes on workers raising future taxpayers – taxes which diminish the future supply of what Adam Smith described as useful labour.
Demographic and fiscal implosion go together as the history of the Late Roman Empire shows.