Your location:

Liam Fox is right – CGT is a bad tax for the economy

    When commentators point out a lack of Lib Dem ‘wins’ in the Coalition government, remind them of the Capital Gains Tax hike undertaken in 2010. Sure, the ‘progressive’ party wanted to increase it further to replicate the income tax bands, but at the time the hike for higher rate income tax payers from 18 per cent to 28 per cent was still a significant increase, especially for a tax that optimal tax theory suggests is so damaging for the economy.

    This is important. The debate on CGT in particular has often been framed purely on the likely effects on revenue, as it is a tax where the peak of the Laffer curve (or the revenue maximising rate) is thought to be at a low rate.

    In a CPS report released yesterday, Lord Flight and Oliver Latham engage in a thorough review of the economic literature and come to the following conclusions:

    • Almost all conventional economic models regard taxing capital as bad: an ideal tax system raises the revenues the government requires in the least distortionary way, with any redistribution done on the spending side. One of the key conclusions of optimal tax theory is that capital gains should not be taxed as part of this. In those models which seek to redistribute equitably suggest this should be done through income taxes. In fact, even in a recent paper by Piketty and Saez (2012) making the case for capital taxation, they argued it should be on bequests rather than investment income. 
    • Capital gains tax leads to inefficient investment allocation because of the ‘lock-in’ effect: when capital gains taxes are high, an individual holding an asset may believe that another asset offers him a higher rate of return, but not dispose of the first asset because the tax liability outweighs the higher return he could earn through buying the second. This can distort individuals’ asset portfolios and lowers expected revenues. For this reason all studies suggests CGT cuts generate a short-term spike in realisations, while in the long-term a substantial portion of the static cost of cutting the tax will be recouped because of the increase in activity.
    • There’s good evidence that cutting CGT boosts growth: because CGT distorts both individual labour market and saving and investment decisions, it does more harm to growth than equivalent taxes on labour or consumption. Mankiw and Weinzierl (2006), for example, find that the extra growth following cuts in taxes on investment income is sufficient to cover 40% to 70% of the static cost of the tax cut.
    • The Government recognises CGT is distortive, hence the number of exemptions: the Government already recognises that CGT is a distortionary tax, hence the huge range of exemptions including private residence relief, business asset roll-over relief, entrepreneurs’ relief, incorporation relief, stocks held in ISAs, gambling winnings, gilt edged securities.

    The conclusions from the literature are clear. CGT should be kept low for reasons of economic efficiency. Cuts in CGT rates from the current level need not have substantial budgetary implications. And the overall optimal rate is often well below the revenue maximising rate.

    Nevertheless, the Government’s declared rationale when increasing the rate of CGT was to ameliorate tax avoidance (i.e. get the rate closer towards the top rate of income tax in order to reduce the incentive to reclassify income as gains) and set the rate such that it maximised overall revenue. They suggested that this entailed balancing off two effects: the loss of revenue due to the lock-in effect, and the increase in revenue arising from lowering tax reclassification.

    The problem for the Treasury now is that it cannot claim the 28% rate maximises revenue when the top rate of income tax was 50% without acknowledging that we will be on the wrong side of the Laffer curve once the top rate falls to 45% in 2013. After the top rate of tax is cut, the difference between the tax rate on income and gains will fall and so the incentive to avoid income tax will be reduced. This implies that if the two effects were equal before, the lock-in effect will be larger now because the cost of reclassification has fallen – so the 28% rate should be cut if revenues are to be maximised.

    Now, of course, the Treasury are likely to be reluctant to reduce the rate, not least because the opposition parties will label it a tax cut for a rich (though bizarrely many support QE despite the distributional impacts). This is quite incredible, particularly because the 18% rate was inherited from the Labour government!

    Nevertheless, it makes absolutely sense to continue with a tax rate which loses the Government potential revenues.

    The Government should therefore use all of this evidence to set out short-term, medium-term and long-term ambitions.

    • Short-term: cut the top rate of CGT so that it is at the very least at the new revenue maximising rate.
    • Medium-term: set out a flat rate of <20%, a level at which the economic literature suggests is less distortionary, and seek to eliminate the taxing of ‘paper gains’ by indexation.
    • Long-term: as part of a fundamental reform of the tax system, think about abolishing CGT altogether.

    Ryan joined the Centre for Policy Studies in January 2011, having previously worked for a year at the economic consultancy firm Frontier Economics.

    Centre for Policy Studies will not publish your email address or share it with anyone.

    Please note, for security reasons we read all comments before publishing.


    It doesn't add up... It doesn't add up... - About 2936 days ago

    You omit to consider the lack of inflation indexation relief (or even the use of tapering as a proxy). As presently constructed, CGT is a swingeing wealth tax on the realisation of longer held assets that can be avoided by borrowing against the assets as collateral (e.g. to fund nursing home fees).

    You also omit consideration of relief for losses, which mitigate the effects when inflation is low and asset prices are weakly correlated, but is effectively not available when asset price correlation is high.

    Of course, some anti-avoidance provision to prevent transmutation of income to capital gain is essential unless you also propose replacing income tax with expenditure taxes.

    View Comments (1)
    Comment on This

    Centre for Policy Studies will not publish your email address or share it with anyone.

    Please note, for security reasons we read all comments before publishing.

    Ryan Bourne - About 2936 days ago

    Almost all of these things are considered in the paper.