Against capital gains taxes

By Eric Crampton 05/06/2013

The OECD having recommended that NZ adopt a capital gains tax, it’s worth reviewing the case against them.

First, here’s the OECD:

New Zealand belongs to a group of five OECD countries with particularly high pre-tax capital-income inequality (Figure 13). As much of this income, especially at the top levels, takes the form of capital gains, the lack of a capital gains tax in New Zealand exacerbates inequality (by reducing the redistributive power of taxation). It also reinforces a bias toward speculative housing investments and undermines housing affordability, as argued in the 2011 Survey. 

The OECD report also seems to reckon that capital gains taxes, along with other taxes and changes to Superannuation, could help debt issues when Superannuation starts getting rather expensive.

Where to begin.

First, the OECD is entirely right that NZ should be moving to increase the age of superannuation eligibility. The Productivity Commission said so, anybody economically sensible has said so, and even Labour’s in favour of it. The only thing that seems to be holding it back is that Key had promised in 2008 not to do it and didn’t change his mind in the last election.

Second, the OECD could be right about land taxes. As part of a revenue-neutral shift away from income taxation, it would be a really nice move. My only concern, and it is the one that would keep me from pushing the button for that particular tax shift, is that equilibrium overall tax rates are likely to wind up much higher as consequence. Open up another margin for taxation and the government will wind up getting bigger over time. Even if it’s revenue neutral now, it won’t be the next time a Labour/Green finance minister decides to go after the ‘rich pricks’ by reinstituting a 39% top marginal tax rate while keeping the land tax.

But on capital gains, well, we disagree.

First, it’s hard to make the case that the absence of a capital gains tax distorts investment towards housing. Maybe you could argue that we’ve a distortion such that firms have incentive to avoid distributing revenues as dividends and that individuals have incentive to hold shares in firms that follow such strategies rather than interest-bearing assets, but that doesn’t make a case for a housing-specific distortion – especially as the IRD has been getting more vigilant about individuals flipping houses. Buying a house, doing it up, and re-selling it at a profit will draw income tax: the gain is taxable income from your labour in fixing and marketing the house. I can’t see how we get a distortion towards housing rather than towards a broad set of appreciating capital assets. We certainly have problems around housing affordability. The first order problem is Councils’ restrictions on the supply of zoned land. Sort that one out, and a lot of the second order problems, like inefficiencies of scale in construction, also start going away.

Second, capital income is already taxed when it is spent: we have a 15% GST. The more we are able to shift from income to consumption taxes, with offsetting transfers to those on lower income if you like, the better.

Third, as Seamus pointed out two years ago, we need to compare the relative efficiencies of the different available tax instruments. Taxes on capital income are more distortionary than taxes on labour income, and even worse when capital gain taxes tend not to be inflation-indexed; real tax rates on capital income then easily wind up being higher than taxes on labour income. And, there’s a bit of a mess in deciding how to treat realised versus unrealised gains – you’re basically there choosing among rather bad consequences. Read Seamus’s whole post.

Fourth, the story required for the absence of a capital-gains tax to distort choices between productive investments and some kind of unproductive investment (basically, purchasing some asset that appreciates in value over time) is especially convoluted (another Seamus post…read the whole thing….).

Fifth, Seamus noted that while you can hang a case for a capital-gains tax on an argument Samuelson made rather a while back, Samuelson’s argument shows that you need a capital-gains tax to avoid the problem of distorted choices among assets whose payoffs are more than 25 years into the future; nearer-term payouts aren’t affected by that kind of distortion. Seamus also hit on a couple of other potential objections.

Want to increase the redistributive potential of the tax system? Increase the GST while increasing income-based transfers to the poor. Why muck up incentives to make capital investments?