Yesterday evening, I had the pleasure of listening to the American economist Art Laffer at the IEA. For those who haven’t heard him before, he’s a lively and charismatic speaker, and the anecdotes he tells from years as an advisor to Presidents Reagan and Clinton as well as Prime Ministers Thatcher and Blair add a human touch to the advice he brings.
Dr Laffer’s speech was a microcosm of his impact on policy. He has an ability to take economic concepts and reduce them to brutal simplicity through examples and logic. This ability is, of course, what led to the derivation of what we now know as the Laffer curve – the simple observation that increasing any rate of taxation beyond a certain limit is likely to be revenue diminishing because of the changed behaviour of the taxpayers.
The impact of his supply-side theory of taxation cannot be understated. Across the developed world, countries have followed the example of Reagan and Thatcher in slashing taxes on profits and income. This perhaps explains why opponents of economic liberalisation seek to deride or caricature his theories – he’s become a victim of his own success. Not only do they falsely claim that Laffer believes all tax cuts are self-financing, but some go further and state that no economists believe in the Laffer concept – an utterly laughable claim (especially when you consider it was observed by Keynes).
Let’s set out exactly what the Laffer curve says about tax rates.
If you tax someone at 0% you get no revenue because you aren’t actually taxing them. If you tax someone at 100%, you get no revenue because nobody has any incentive to work or report tax. There must, therefore, be a rate in between 0% and 100% at which revenue is maximised.
In between the two extremes every increase in a tax rate has two effects. First, everyone who remains paying the tax now pays more.
But there is a second, more nuanced effect. An increase in the tax rate diminishes the incentive to work or produce, and/or encourages people to avoid or evade paying the tax. If the scale of this second effect exceeds the first effect, then we are beyond the revenue maximising rate – or to the right of the peak of the Laffer curve. And the key point here is this: even if the second effect does not exceed the first, and we are not beyond the peak of the Laffer curve, the second effect at least dampens the first effect, meaning that revenues do not increase as much as a static analysis would suggest.
I know of no serious economist who disagrees with this premise. It is basic economic fact. Where we disagree is on where the revenue maximising rate is for each tax.
This is what makes some of the articles in reaction to Laffer’s visit so silly. Take the Comment is Free article written in the Guardian by Michael Burke. He claims that there is no basis for the Laffer curve on a ‘theoretical level’ – linking to a poll of American economists which apparently shows none of them believe in it.
Except it doesn’t show any such thing. The suitably vague first question, which fails to address for whom the tax cuts would come asks whether ‘A cut in federal income tax rates in the US right now would lead to higher GDP within five years than without the tax cut’. Most agree or are uncertain, but in the notes many sensibly note that it depends whether this is financed by deficit spending. The second question simply asks whether this tax cut would be self-financing. Unsurprisingly, most disagree. The US has low rates of tax, and a sole income tax rate cut would unlikely be growth inducing enough to increase revenues.
But note, nothing here has contradicted the Laffer curve concept. In fact, by recognising the growth effects that lower rates of tax can have, the economists have noted that some of the direct loss in revenue from the tax rate cut will be offset from higher revenue due to growth, likely to affect other taxes as well – meaning at least some of the projected static revenue gain will not be realised. All the economists are saying is that for income tax in the US, we are at the moment to the left of the Laffer curve peak.
We know this is not always the case. Take the 50p rate here. A static analysis of the revenue gain from increasing the top tax rate in the UK from 40p to 50p by Richard Murphy suggested the tax could raise over £6 billion. But even the initial Labour government forecast recognised that because of behavioural effects it would only raise £2.7 billion. HMRC’s subsequent analysis for Budget 2012 found the gain was just £100m, and that is when the tax was still thought to be temporary. A permanent rate at 50p is therefore almost certainly likely to be beyond the revenue maximising rate.
There are other examples. As Laffer set out, capital gains tax cuts in the US have almost always led to increased revenues, because of the ‘lock-in’ effect. In 1986, the top rate of tax on long-term gains increased from around 20 per cent to over 30 per cent. There was an initial spike in revenue just before the change as investors sold assets in anticipation of the rate hike. But then revenues plummeted. Policy makers were convinced that government revenues would quickly bounce back until they exceeded pre-reform levels, but revenues continued to fall in real terms for another four years despite good stock market performance. Clearly, the US capital gains tax rate was either to the right of the Laffer curve peak already, or was taken there by the tax hike.
There are, therefore, instances where cutting rates can reduce avoidance and improve incentives to the extent that the tax cuts become self-financing. But Laffer does not claim this is always the case.
Why then do Laffer’s views on tax get misrepresented?
I’d suggest two explanations.
Firstly, because of his name, every time Dr Laffer advocates a tax rate cut people assume he is utilising the theory that we are on the wrong side of the Laffer curve. But, here’s the thing. Dr Laffer is an advocate of limited government. He doesn’t believe production and human endeavour exist to maximise revenues for government, and so sometimes – shock horror – he advocates lower taxes solely on the grounds of their growth effects or because he thinks they are the right thing to do. The revenue maximising rate shouldn’t be confused with the optimum rate. Sometimes he believes government should be smaller to pay for tax rate cuts, and advocates of limited government aren’t popular with many left-wingers.
Secondly, he uses the Laffer concept to inform his recommendations for tax policy. In particular, he notes that the richest tend to find it easiest to avoid and evade tax, and in the extreme case can move to different tax jurisdictions. Since this a fact of life, he suggests that high income taxes on the rich are often to the right of the Laffer curve peak and it is best to try to increase overall revenues by having a competitive rate which keeps people here – not only earning income, but also buying goods and services, investing, flying and subsequently paying all manner of other taxes. He recognises that complex tax systems with hugely differential rates make this less likely to occur, and destroy incentives for work and production across the income scale. So he advocates a flat tax: with the lowest possible rate across the broadest possible base to provide a level playing field and all but eliminate avoidance and avoidance, and reduce the size of the underground economy.
Now, there are reasons why people might object to small government and flat taxes. Many do. But to use opposition to these to claim that the whole Laffer curve concept itself is bogus is just disingenuous and wrong.