The debate around the capital gains tax (CGT) ‘discount’ is really heating up due to budgetary pressures, and arguments around ‘intergenerational equity’ and housing affordability. But the so-called ‘discount’ is not a concession, rather it is an ‘adjustment’ to approximate gains arising from inflation that should not be taxed.
Prior to 1999, the cost base of an asset was indexed to inflation, such that only real gains were taxed. Post-1999, a 50% CGT ‘discount’ applied to nominal gains as a blunt approximation to allow for inflation.
Rhetoric in the media amid calls for a reduction in the CGT discount, centres on the supposed advantage it gives to property investors. So I have dug deeper with some numbers, to work through what it all really means and hopefully dispel a few myths along the way.
Many argue that the current level of discount at 50% applied to assets held for more than twelve months is too generous if the asset is sold after only a short period, say a year. But consider a property purchased a year ago that grew in value by 5%, with 2.5% inflation over that period.
Under the pre-1999 approach, setting buy/sell costs aside, the real gain would be calculated as:
(1.05 x purchase price) – (1.025 x purchase price) = 2.5% of the purchase price.
The same result as a 50% discount applied to the nominal gain of 5%. So that’s not generous.
The argument also goes that the 50% discount is however, more justifiable if assets are held for longer durations. Using the same assumptions, but suppose the property is held for ten years before disposal.
Pre-1999 real gain = (1.0510 - 1.02510) x purchase price = 34.9%.
Total nominal gain = (1.0510 – 1) x purchase price = 62.9%.
Inflationary gain proportion = 1 – (34.9% / 62.9%) = 44.5%.
That is, full indexation implies a discount of 44.5% required to remove the inflationary component of the nominal gain, leaving the real gain of 34.9%.
However, applying the current 50% discount would pare the net gain back to 50% x 62.9% = 31.5%. Which in this instance would be more generous.
In fact, as long as the nominal growth rate is greater than the inflation rate, the real gain proportion of total gains grows over time and the inflation share falls, such that the required discount rate falls. That is, a fixed 50% discount becomes progressively more favourable with duration.
This contradicts the ‘property flipper’ argument that says short-term sellers benefit unfairly because they get the same 50% discount as long-term investors. It is at odds with peoples’ intuition because they conflate the dollar gain with the composition of that gain (inflation vs real).
The example also reveals that from end year 1 to end year 10, the actual discount rate required such that only actual gains are taxed, has moved barely five percentage points from 50% to 44.5%.
If the inflation assumption was instead 2% with 5% asset growth, the corresponding implied discounts required at end year 1, and end year 10 would be 40% and 34.8% respectively.
And for an inflation rate of 3%, those discounts come in at 60% and 54.7%.
Some more observations:
- Whether the property is held for one year or ten years while holding inflation constant, the real versus nominal splits change slowly over time, leading to a gradual decline in the discount rate from year to year. Duration matters unintuitively little.
- It is the inflation rate that shifts the dial. A change in inflation from 2% to 3% moves the implied discount from 35% to 55% when sold after ten years. That is, the proportion of real gains to total gains is the more significant driver of the discount rate required to remove inflation gains.
- A 50% discount assumption is an attempt to balance out actual inflation outcomes. It implies that inflation accounts for about half of total nominal growth over the long-term. An assumption broadly in line with headline inflation of around 2-4% since the late 1990s, and nominal asset growth of 6-8%.
- If however, inflation is persistently lower such that the real gain proportion is greater than 50%, the discount over-adjusts (is generous) because a smaller discount is actually required to counter the inflationary gains. The converse holds if inflation is higher.
Advocates who call for a reduction in the CGT discount on the grounds of ‘intergenerational equity’, argue that lowering it increases fairness, when it really moves in the direction of taxing inflation.
Those people ignore the notion that the discount is a crude inflation assumption. And that such an adjustment by design will sometimes result in under-taxing real gains, and sometimes over-taxing.
The discount is not generosity favouring long-term holdings, but rather is a consequence of accounting for actual inflation with a fixed proxy.
And often overlooked is that a CGT discount also partially offsets distortions implicit when tax is triggered by a one-off event: the accumulation of multi-year gains into a single tax year, and the ‘lock-in effect’, or choice not to sell, that arises because gains are taxed only on sale.
Therefore changing the discount is not correcting for ‘intergenerational equity’.
Similar, is the housing speculation argument that says that reducing the discount will deter investors. But the discount is not rewarding speculation, it is not a behavioural concession. Its purpose is to avoid taxing inflation.
Changing the discount rate to address housing affordability would be misplaced policy, when addressing the supply side of housing would be the way to tackle housing affordability. Surely the ‘wisdom’ of increasing tax on something you need more of, is counter-intuitive.
Finally, proponents of a CGT discount reduction will point to Treasury’s Tax Expenditures and Insights Statement (TEIS) for justification, citing claims like the 50% discount cost the budget around $19.7 billion in 2024-25.
Such claims however, are misleading because the TEIS uses a benchmark where nominal capital gains are fully taxable when realised. Meaning the ‘revenue foregone’ makes no allowance for inflation, such that calculated ‘costs’ are significantly overstated. In any case, not taking more of taxpayers’ money, is not a ‘cost’ to the budget.
This whole debate really gets back to terminology. Call something a ‘discount’ and people will assume that recipients of such are getting a leg up. Why not remove that perception with the simple remedy of returning to explicit CPI indexation of the cost base? That would remove both the ‘discount’ rhetoric, and any ambiguity around what is being adjusted.
Tony Dillon is a freelance writer and former actuary. This article is general information and does not consider the circumstances of any investor.