Are capital gains taxes fair?🍋

NZ lacks a specific capital gains tax. But introducing one may not increase fairness.

Unlike most other Western countries, NZ does not tax most capital gains.¹ As a result, there have been recent calls for general implementation of a capital gains tax (CGT) by some politicians and business leaders, emphasising the unfairness of capital gains not being taxed whilst other forms of income are.

This implies that people who own assets generating tax-free capital gains have an advantage over others. Many tax experts share this view. For example, Burman and White argued that taxing capital gains enhances equity because they are a form of income and other forms of income are taxed, while the Tax Working Group argued that “…taxing capital gains would improve the fairness of the tax system by reducing inconsistency in the treatment of income, no matter how it is earned.”² These views suffer from three fundamental problems, which I outline in this post.³

Problem 1: Double taxation

Take a capital gain that arises solely from the retention of earnings that have already been taxed as company profits. Taxing this gain would be double taxation of the retained earnings — once as profit, and again as a capital gain.

For example, suppose that a person A pays $10m to purchase all the shares in company B. Over the next year B earns a pre-tax profit of $1m, which is taxed at 30%,⁴ leaving $0.7m, which B retains as cash. B’s total value is now $10.7 million. Next, A sells the shares to C for $10.7m. A has made a capital gain of $0.7m, which attracts a CGT (at the 30% rate) of $0.21m. The $1m was only earned once, yet it has been double-taxed, first as profit of B and then a second time as capital gains of A, for an effective tax rate of 51%.

Fiordland coastline Dave Heatley

Problem 2: Semi-strong market efficiency

Even if capital gains on an asset do not arise from retained earnings and therefore the absence of CGT constitutes favourable tax treatment of the asset, this favourable tax treatment is impounded into the price paid for the asset. Therefore the recipient of the capital gains does not benefit from the absence of CGT. For example, suppose that a person purchases farmland and leases it out for grazing, and this currently generates revenue of $10m per year, and none of these lease revenues are retained for further investment into the property. Suppose also that there is no CPI inflation, and therefore all capital gains or losses are real.⁵ Since real GDP is expected to grow, and land is in fixed supply, its value and the grazing revenue are expected to grow, at a rate of 2% per year. Real capital gains are therefore expected, but they do not arise from retention of some of the asset’s earnings.⁶

Suppose further that investors are risk neutral, i.e., they require no compensation for risk. In addition, the government borrowing rate (the “risk-free” rate) is 5% per year and the personal tax rate is 30% for all investors. These assumptions imply that all assets must be priced so that their expected rate of return after all taxes is equal to that on government bonds, and the latter rate is 5% (1 – 0.3) = 3.5%. In addition, suppose that the farmland in question is owned via a company, whose lease revenues are taxed at the corporate rate of 30% and the after-tax revenues are paid as dividends to the landowner. Furthermore, dividend imputation operates, with the result that the company taxes that are paid generate imputation credits that fully offset any personal tax on the dividends.⁷ So, the discount rate on the after-company tax cash flows from the land is 3.5% per year. The value of this land would then be $467m as follows:

\(V=\frac{$10m\left(1-0.3\right)}{1.035}+\frac{$10m\left(1.02\right)\left(1-0.3\right)}{\left(1.035\right)^2}+\ldots=\frac{$10m\left(1-0.3\right)}{.035-.02}=$467m\)

A person buying the land at this price then expects to receive a cash yield (after company tax) on their investment of $10m(1 – 0.3)/$467m = 1.5% in the first year, and a tax-free capital gain of 2%, for a total of 3.5%. Their after-tax expected rate of return therefore matches that on the risk-free asset (and every other asset) despite part of it being tax free. So, the fact that part of their return is tax free gives no benefit to them because the price they pay for the asset is higher to reflect that tax feature.

If capital gains were taxed, the cost of capital would be higher to reflect that fact. In particular, it would be higher by the product of the CGT rate and the expected rate of capital gain (2%).⁸ So, if capital gains were taxed at 30%, the cost of equity would be:

\(k=5\%\left(1-0.3\right)+2\%\left(0.3\right)=4.1\%\)

Using this in substitution for the earlier cost of equity, the value of the land would then be $333m:

\(V=\frac{$10m(1-0.3)}{1.041}+\frac{$10m(1.02)(1-0.3)}{{(1.041)}^2}+\ldots=\frac{$10m\left(1-0.3\right)}{.041-.02}=$333m\)

A person buying the land at this price then expects to receive a cash yield (after company tax) on their investment of $10m(1 – 0.3)/$333m = 2.1% in the first year, and an after-tax capital gain of 2%(1 – 0.3) = 1.4%, for a total of 3.5%. Again, their expected rate of return matches that on the risk-free asset (and every other asset). So, asset owners gain no advantage nor suffer any disadvantage from the tax treatment of their assets, so long as the tax treatment predates their purchase.⁹

This analysis assumes that market prices already reflect all public information, including tax rules. Economists call this the semi-strong form of the efficient markets hypothesis. It is supported by an immense body of empirical work.¹⁰

Problem 3: Transitional unfairness

Consider the example above, with no CGT and therefore a value for the asset of $467m. If a CGT were introduced, the asset value would immediately fall to $333m, causing a capital loss for whoever owns the asset at that moment.¹¹ Subsequent owners of the asset would not be affected by the introduction of the tax, despite them also paying it, because the lower price they paid for the asset would compensate them for these taxes. Thus, the only owners who would be adversely affected by the tax are those holding the asset at the time the tax is introduced. This is highly inequitable, and especially so since those holding the asset at this time never benefited from the earlier absence of the tax.¹² So, introduction of the tax creates a new inequity, and without eliminating any existing unfairness.

One might then wonder who has benefited from the absence of CGT on farmland. Most land purchases in NZ ultimately derive from settlers purchasing it from the Government, who in turn acquired it from Māori. The prices in these transactions between settlers and the government reflected the absence of CGT, i.e., higher than they would otherwise have been. So it was actually the Government that benefited from the absence of a CGT.

Conclusion

Many capital gains simply reflect profits that have already been taxed once and retained. Charging CGT on these would amount to taxing the same earnings twice — first as company profits, then again as capital gains. Even when capital gains arise from other factors, their tax treatment is already built into the price of the asset. Buyers pay more for tax-favoured assets, so they don’t actually enjoy any benefit from the favorable tax treatment. Introducing a CGT would therefore hurt only current owners, whose asset values would drop overnight. New buyers wouldn’t be worse off. A CGT would create new inequities without fixing old ones.

By Martin Lally

1. Does New Zealand Have a Capital Gains Tax (CGT)? contains a list of capital gains that are treated as income in NZ, and thus taxed at the taxpayer’s marginal tax rate. This includes gold, cryptocurrencies, many rental properties, and some shares.

2. See page 21 of Burman, L. & D. White (2009) Taxing Capital Gains in New Zealand: Assessment and Recommendations, and page 60 of Tax Working Group (2019) Future of Tax: Final Report Volume 1.

3. A more detailed treatment of this topic appears in Lally, M. (2025). The Fairness Case for Capital Gains Tax in New Zealand: A Financial Economics Perspective.

4. The NZ company tax rate is 28%. I use 30% in my examples to make the arithmetic simpler.

5. This eliminates the issue of whether to tax the CPI component of capital gains, allowing me to focus on the key issue.

6. The principles in this section apply equally to the land component of residential property that is leased. However, the value of such property involves both land and buildings, and buildings are not subject to their value growing at the same rate as land. So, to simplify the analysis, farmland is used in the example.

7. These additional assumptions simplify the analysis without altering the principal conclusions.

8. This assumes the tax is applied annually, regardless of whether gains are realised within that period. If it applies only to realised gains, when the sale occurs, the adjustment to the cost of equity would be smaller. Allowing for this does not alter the conclusion that the owner gains no benefit from the absence of CGT.

9. In addition to the existence of a CGT raising the cost of capital, it might also raise the lease revenues. If so, the drop in the land value would be less but it would still be true that a person purchasing at this price would still expect to receive a rate of return after all taxes of 3.5%.

10. For example, see Ch. 11 of Copeland, T., J. Weston, & K. Shastri (2005) Financial Theory and Corporate Policy, 4^th edition, Pearson Addison-Wesley, Boston.

11. By reducing the expected future post-tax payoffs from (now) taxable assets, a CGT announcement lowers the market price of those assets. Should a policy analyst fail to take this into account, they could overestimate future CGT revenue. For example, say the country’s taxable assets are $100bn. If realised capital gains were assumed to average 2% of asset values per year, the expected revenue from a fully operational 30% CGT would be $0.6b per year (slowly growing thereafter). But if the announcement of a CGT took say 25% off the market price of existing assets, the expected revenue would also drop by 25%.

12. If the introduction of the tax raised the lease revenues, the reduction in the asset value would be less severe and therefore the burden of the tax would fall on those owning the asset at the introduction of the tax and those leasing the land, and the latter might flow through to those consuming the outputs from the land. However, those purchasing the asset after the introduction of the tax would still not bear any of the burden.