Revaluation and reform of council tax in Scotland: design considerations and potential impacts

5. Rents and property values

This section of the report considers potential impacts on rents and property values. We begin by explaining the evidence – theoretical and empirical – on how property taxes end up leading to changes in rents and property values. We then examine the potential impacts of the different modelled reform options on property values in Scotland, including breakdowns by property value and the household income of owners.

5.1 Evidence on impacts on rents and values

Because rents and property prices are determined by the supply of and demand for housing, we would expect changes in council tax bills to be reflected over time in properties’ market rents and values: changing the cost of occupying property changes the demand for it. Holding other factors fixed (such as the level of local public expenditure), we would expect market rents and prices to rise for properties that see a reduction in their council tax bills as a result of revaluation and reform, and fall for properties that see an increase in their council tax bills. The reflection of future tax bills in current property prices is a process known as capitalisation.

Economic theory suggests that tax changes will be most strongly reflected in rents and property values where housing supply is relatively unresponsive to property values (Oates, 1969), as it is in the UK (Caldera Sánchez and Johansson, 2011; Drayton, Levell and Sturrock, 2024). This is because in these circumstances, any changes in property supply in response to changes in prices will be smaller, and so offset less of the initial change in property values. In places where council tax bills were reduced, for example, the reduced cost of occupying housing would push up demand and therefore property prices; those higher prices would strengthen the incentive for developers to build in that area, but the evidence suggests that any resulting increase in the supply of properties would be relatively small, and therefore insufficient to bring values and rents back down.

Empirical studies from the UK also find that for a given level of taxation, expenditure on, and the quality of, local public services – the corollary to local taxes – is highly capitalised into property prices (Hilber, Lyytikäinen and Vermeulen, 2011; Gibbons and Machin, 2008). Many studies on other countries have also found that property taxes are reflected almost pound-for-pound in local rents and property values (Capozza, Green and Hendershott, 1996; Palmon and Smith, 1998; Hilber, 2015; Høj, Jørgensen and Schou, 2018).There is therefore strong empirical backing for the theoretical prediction that property prices would be affected by changes to property taxes such as council tax.

It is important to note that the theory of capitalisation implies that owner-occupiers whose bills rise do not suffer a ‘double hit’ from both the increase in their council tax bill and the fall in the value of their property (Adam et al, 2020a, 2020b). On the one hand, if they continue living in the property indefinitely, they lose or gain as a result of the change in their tax bill by paying a higher or lower bill than in the absence of the reform. On the other hand, if they sell and move immediately, they lose or gain as a result of the change in their property value, but no longer have to pay council tax on the property. They pay the higher/lower council tax only for as long as they live in the property; they crystallise the decrease/increase in property value only when they sell it. An owner’s loss from a reduced property value is in effect deferred until it is crystallised when the property is sold; deferral of that loss is valuable, and the additional stream of council tax they must pay until they sell it can be thought of as the ‘price’ paid for that deferral.

The total value of their loss is the same whether they sell immediately or only in the distant future; all that changes is how much of their loss is experienced in changed council tax payments and how much in a more/less distant fall in the proceeds from sale – (obviously the same is true for a rise in value caused by reduced council tax bills). However, capitalisation does mean that it is the owner of a property at the time of revaluation and reform who loses or gains: future purchasers will pay less for the property if the tax bill associated with it is higher, and vice versa.

In the case of rental property, the immediate impact of a change in council tax liability will be felt by the tenants who have to pay it (unless the rental contract specifies that the landlord will pay the council tax). Over time, rents will adjust as described above, passing the impact onto landlords; these adjustments would typically be associated with renewal of the tenancy or a change of tenants, though the dynamics of market rent adjustments can be more complicated than that. Again, however, the theory of capitalisation implies that the value of the landlord’s property will change immediately, reflecting the change in expected future rental income associated with the property (or the corresponding change in council tax liability if the property were to be bought for owner-occupation).

5.2 Potential impacts of reform options

We consider the potential impacts on property values assuming changes in council tax are fully capitalised into values. However, how large these changes would be is uncertain, depending not just on how council tax bills change, but also on how people value changes in expected future tax bills. The more they value those changes (that is, the less they ‘discount’ changes in future bills), the larger the impact any given change in council tax bills will have on property values today. Because there is uncertainty about just how much people discount the future, we therefore estimate property value impacts under three different scenarios for the market ‘discount rate’. Our central scenario assumes a 3% real discount rate – meaning, approximately, that after adjusting for inflation, £1 in a year’s time is valued at around 97p today – broadly consistent with long-term real interest rates on UK government borrowing, reflecting the rate at which market participants value additional money at different points in time. Low and high scenarios assume 1% and 5% real discount rates, respectively, reflecting the fact that people may value future tax payments more or less than under our central scenario.

Effects by property value

Figure 5.1 shows the estimated effects on property values of the different reform options assuming a 3% real discount rate. The horizontal axis shows Q3 2024 property values, while the vertical axis shows the estimated values after accounting for capitalisation of changes in tax bills into property values.

Across the bulk of the property value distribution, estimated changes in values would be relatively modest for the pure revaluation and 12-band and 14-band systems modelled. For example, at the 25^th percentile of the distribution (around £100,000), estimated values would increase by around 1% under both a pure revaluation and the 12-band differentiated system, and around 3–4% under the 12-band and 14-band less regressive systems.^[10] Estimated values would increase a little because, on average, properties at this point in the distribution would be estimated to see a fall in their net tax bill, especially under less regressive systems. For median-value properties (around £155,000), estimated values increase by an average of 0–1% under these systems, reflecting small falls in average net tax bills. At the 75^th percentile (around £250,000), estimated values change little, reflecting little change in average net tax bills.

Note: The figure shows 100 groups of current property values. It omits the two highest groups with values of around £627,456 and £997,395. Pure reval = pure revaluation; 12-band diff. = 12-band differentiated; 12-band LR = 12-band less regressive; 14-band = 14-band less regressive; continuous = continuous proportional.

Source: Authors’ calculations using Understanding Society waves 8-10 and 14 and TAXBEN, the IFS tax and benefit microsimulation model.

We estimate that changes would be larger at the very bottom and top of the distribution of property values. For example, for properties worth £60,000, while a pure revaluation would be expected to have little impact on values (most such properties are already in Band A), the 12-band systems would cause an estimated increase in value of around 6%, and the 14-band system an estimated increase of around 8%. Conversely, we estimate that a property worth £500,000 would fall in value by around 2% under a pure revaluation and 3–5% under the 12-band and 14-band systems, as average net council tax bills increase.

Changes would also be larger under a continuous proportional system, reflecting the larger estimated decreases and increases in net tax bills under this system. For example, if fully capitalised at a discount rate of 3%, we estimate that properties worth £60,000 would see an increase in value averaging 16%, while those worth £500,000 would see a decrease in value averaging 7%.

Figure 5.2 shows that the change in property values driven by a given change in tax bill depends crucially on the discount rate assumed. Changes under a 1% discount rate would be three times as large as under our baseline assumption of a 3% discount rate. For example, a £100 reduction in annual net council tax bill would increase the value of a property by £10,000 if discounted at 1% a year, versus £3,300 if discounted at 3%. Conversely, a 5% discount rate would imply changes in property values 40% smaller than with a 3% discount rate: just a £2,000 change in property value for the same £100 change in annual net council tax bill, for instance.

This means the aforementioned figures can be multiplied by 3 or by 0.6 to calculate the average estimated change in values under a 1% or 5% real discount rate, respectively. Figure 5.3 illustrates this for the 12-band less regressive system. For example, we estimate that properties currently worth £60,000 would on average rise in value by 18.5% if the discount rate is 1%, rise by 6.2% if the discount rate is 3%, or rise by 3.7% if the discount rate is 5%. We estimate that properties worth £169,000 would on average rise in value by 3.9% with a 1% discount rate, by 1.3% with a 3% discount rate, or by 0.8% with a 5% discount rate. We estimate that properties worth £500,000 would on average fall in value by 15.7% with a 1% discount rate, by 5.2% with a 3% discount rate and by 3.1% with a 5% discount rate. Thus, while evidence is strongly suggestive of an impact on property values, there is uncertainty about the scale of these impacts.

Source: Authors’ calculations.

Note: The figure shows 100 groups of current property values. It omits the two highest groups with values of around £627,456 and £997,395.

Source: Authors’ calculations using Understanding Society waves 8-10 and 14 and TAXBEN, the IFS tax and benefit microsimulation model.

Effects by household income

Because lower-income households tend to live in less valuable properties, and higher-income households tend to live in more valuable properties, average changes in property values as a result of revaluation and reform of council tax would differ systematically across the income distribution.^[11]

Note: Assumes full take-up of CTRS. Owner-occupier households are allocated to quintiles based on income measured after taxes and benefits but before housing costs are deducted, and are adjusted for household size and composition using the modified OECD equivalence scale. Pure reval = pure revaluation; 12-band diff. = 12-band differentiated; 12-band LR = 12-band less regressive; 14-band = 14-band less regressive; continuous = continuous proportional.

Source: Authors’ calculations using Understanding Society waves 8-10 and 14 and TAXBEN, the IFS tax and benefit microsimulation model.

Figure 5.4 shows estimated changes in average property values by household income quintile group among owner-occupiers and for rental properties, under the baseline 3% real discount rate assumption. We estimate that a pure revaluation and the 12-band differentiated system would have little effect on average property values across the income distribution – reflecting the fact that it would have little effect on average bills across the income distribution (see Figure 4.3). In contrast, a continuous proportional system is estimated to increase average values for lower-income owner-occupiers (by 2% for those in second-lowest fifth of the income distribution), as well as increase the values of rental properties (2.5% for privately rented properties and 4% for social rented properties). In contrast, we estimate that average values would fall by an average of 3.6% for owner-occupiers in the top fifth of the income distribution under a continuous proportional system.

Estimated effects for the 12-band and 14-band less regressive systems would lie between the benchmark pure revaluation and continuous proportional systems. For example, we estimate that the second-lowest-income quintile of owner-occupiers would see an increase in average property values of 0.8%, and owners of private and social rental properties increases of 1.2% to 1.6%. Conversely, we estimate that average property values for owner occupiers in the top fifth of the income distribution would fall by around 1.4%.

Figure 5.5 shows how estimated effects vary depending on the real discount rate assumed. Under a 1% real discount rate, for example, we estimate that the impact of the 12-band less regressive system would be similar to the impact of the continuous proportional system under a 3% real-discount rate, with average property values estimated to increase by around 1–2% for low-income owner-occupiers and 3–4% for rental properties, and estimated to fall by around 4% for high-income owner-occupiers. Estimated impacts under a 5% real discount rate would be much smaller.

Note: Assumes full take-up of CTRS. Households are allocated to quintiles based on income measured after taxes and benefits but before housing costs are deducted, and are adjusted for household size and composition using the modified OECD equivalence scale.

Source: Authors’ calculations using Understanding Society waves 8-10 and 14 and TAXBEN, the IFS tax and benefit microsimulation model.