Will Hutton plumps for Mirrlees

Makes me wonder about my own support for the Mirrlees Review. For Willy is indeed a normally perfect indicator of what not to do. But perhaps this is his stopped clock moment.

In fact, the outlines of what such a reforming chancellor might do were recently set out in an extraordinary review of the tax system led by Sir James Mirrlees of Cambridge University for the Institute for Fiscal Studies. Published last autumn after nearly five years\’ work, it has received far less attention than it deserves since it is the most comprehensive and devastating indictment of the current situation.

This is actually a very useful description of politics as it actually works. Not as it ought to, but as it does.

We have two sets of alternative proposals about how we ought to reform the tax system. One has been composed by a Nobel Laureate (one whose Nobel was awarded largely for his studies of tax systems), has taken 5 years to complete and is comprehensive.

The other is cobbled together by a retired accountant from Wandsworth who seems to have terrible difficulty in understanding the most basic points about the economics of tax systems. The existence of the Laffer Curve for example, the incidence of taxation. Indeed, he\’s been vocal in support of a transactions tax something which out Laureate has pointed out is a very silly method of taxation indeed.

And which set of proposals gain traction with the TUC, the BBC, The Guardian, with he left generally?

Quite, it\’s something of an indictment of how politics actually works, isn\’t it?

25 comments on “Will Hutton plumps for Mirrlees

  1. The Mirrlees proposals are eminently sound and reasonable. However is there any precedent for a mature economy instituting such a radical and wide-ranging change of their entire taxation system ?

  2. Ah, but Mirrlees is not a green eyed, ‘what’s yours is mine, what’s mine’s my own’ type of person. Nor is he paid by the forgoing to cloak their hate in a veneer of reason. I suspect that is because R.M. does not, in fact, understand their loathing that he has to mangle his economics to fit their prejudices.
    As you regularly point out, R.M.’s actions speak louder than words in the way he has used perfectly legal means to reduce his tax burden.

  3. The existence of the Laffer Curve for example, the incidence of taxation.

    Which brings us back to this problem that there doesn’t seem, from basic logic or indeed statistics, to be a Laffer Curve. It is certainly the case that tax revenues will vary depending on the disposition of a complex tax system. But a Laffer Curve? Hmmm.

    The logic behind the LC is this:

    (a) if tax is 0%, revenue is zero.

    (b) if tax is 100%, revenue is zero.

    (c) if tax is between those figures, revenue is greater than zero. Ergo, there is a humpy curve from 0 to 100. QED.

    The problem with this is that a tax rate of 100% doesn’t produce revenue of zero. It produces revenue of 100%. It simply describes a fully socialist economy. Everyone gets a pay slip saying something like: “wage: £1000. Tax £1000”, and all services are provided by the State out of the tax revenues. All economic activity occurs in the Public Sector.

    So there’s your problem. The right hand end of the putative curve isn’t anchored at 0, it’s anchored at 100; and you can’t do better than that in terms of revenues.

    Bit of a problem, that, then.

  4. Aye, Ian, but of course under the socialism you describe, the tax take would be 100% of bugger all.

    I think the statisticians always warn to look hard at what they call the “initial conditions”.

  5. >It produces revenue of 100%.

    The revenue needs to be described in absolute terms (ie. the actual amount of money raised), not in terms of the percentage that is taken from people.

    You’re right that it won’t be zero in absolute terms either. But it won’t be that high either.

  6. Aye, Ian, but of course under the socialism you describe, the tax take would be 100% of bugger all.

    Nope, it’d be 100% of whatever arbitrary quantity of money the State chooses to slosh around its internal “market”.

    Imagine instead, say a 99% tax rate, that makes it easier to visualise I think. You’ve got 1% going to employees to spend on a few trinkets, but 99% in the Pubic Sector.

    And so on. I’m guessing that the conceptual error being made by the Lafferists is the belief that only the private sector pays taxes even though, in our own and everyone else’s system, public sector employees have a tax bill too. To get around that, imagine a (entirely plausible) system in which all government services are dispensed by government contractors, with “private sector” employees receiving pay slips (but in this extreme case, a Net of zero). The money just sloshes back and forth between the government and the contractors. When it’s going from State to contractor, it’s a payment for services rendered, when it’s going back to the State, it’s a tax. It could even quite happily slosh from contractor to contractor.

    The central point being, that there is a money supply, and all of it is taxed, and revenue is 100%. Not 0%.

  7. Tom

    The revenue needs to be described in absolute terms (ie. the actual amount of money raised), not in terms of the percentage that is taken from people.

    It’s generally described as a percentage of GDP for some inexplicable reason, most likely so far as I can tell so that Lafferists can find two different axes to put on their graphs.

  8. @ Ian B
    If the tax rate was 100% I should stop working.
    If the tax rate was zero, I should pay no tax.
    My effective tax rate was too high under Brown so I did less paid work and more unpaid work than under Major/Clarke and paid *less* tax.
    So, in my personal case the Laffer curve does exist. QED
    Your argument is based on the implausible supposition that no-one cares about whether they get any spending money in exchange for their labour so no-one would ever go on strike to get a pay rise.

  9. If the tax rate was 100% I should stop working.

    Starve yourself to death would you? In protest? Don’t think you would, not when your government contractor employer provides your food, your healthcare, your housing…

    All free, of course.

    Of course the point is not whether such a system would be acceptable to people. I’m sure it wouldn’t. People have revolted over far lower tax rates and general impositions. But this is a thought experiment that doesn’t ask what would trigger a revolution or disobedience, but what the government income from such a system would be. And the answer is, it would be 100%, no 0% (or, a cardinal value of 0 if you prefer), as is claimed.

  10. IanB (#11), if the worker gets food and housing in return for working, then the tax rate can’t be 100% because the worker is still getting something, after tax, in return for his work.

    For the tax rate to be truly 100%, the State would provide the food and housing irrespective of whether or not he does any work.

  11. Richard #12

    He’s not getting anything in return for his work, in a market sense. He’s getting free food and housing, as we get the NHS. The government could provide various impulsions and compulsions to actually go to work, ranging from the ideological to brute force. The particular mechanics don’t matter; this is a thought experiment. I think one way to understand it is, although the worker is “getting something” he isn’t being paid it. Whatever the compulsion to work is, it’s independent of his pay. (Which being entirely taxed by The Courageous State is net zero.)

  12. IanB:

    “He’s not getting anything in return for his work, in a market sense. He’s getting free food and housing, as we get the NHS. ”

    Take away the NHS, you’re not describing socialism, but rather, Alabama circa 1860. They worked, and did get fed and housed.

    Which is why the tax take would be zero at 100% tax rate unless your workers were slaves.

    Even if the tax revenues aren’t zero, do you really think that tax take would be maximised at 100%?

  13. Adrian, whether or not people in such a system are slaves or in some other morally objectionable state isn’t relevant to the discussion. The single issue I am addressing is whether a tax rate of 100% would produce revenue of zero or 100%. The answer is the latter.

    Another way of addressing this; currently that crapulent statistic “GDP” which Lafferist revenues are measured as a percentage of consists of the sum of

    (a) a survey of private sector order books, which hopefully measures private sector production. It doesn’t, but let’s pretend it does for now.

    (b) Public spending.

    plus some sundries. In our 100% tax regime, GDP collapses to purely public spending, as there is no private sector. As a consequence, it is now very clear to see that the 100% tax recoups 100% of GDP, which is equal to public spending (being the same thing).

    That is what matters here. The Laffer Curve postulates tax revenues of zero on the right hand side of the graph. This is demonstrably wrong.

    So, rather than the curve being a bell of some shape from (0,0) to (100,0) with higher values in between, it has to be a line from (0,0) to (100,100), the shape of which is not part of this argument. But it can’t go higher than 100%, that’s tautologically true.


    I would also postulate that were GDP actually a measure in money-units of production, the line would be a straight line. Due however to vagaries of the measurement system, and enormously complex tax systems which allow wise citizens to shift tax liabilities hither and yon, plus interactions with foreign economies, all kinds of actual measured figures are found (e.g. 30:40 or 30:20 etc).

    But that’s not the main point.

  14. Ian, the Laffer curve is meant to describe the effect of tax in a free(ish) society, so the fact that we could be compelled to work in return for nothing doesn’t invalidate the theory.

    And in all the graphs I’ve seen, revenue under the Laffer curve is in $, not % of GDP. So at 100% tax the tax take can be zero dollars. That could still be 100% of GDP, but that isn’t what it’s trying to do.

  15. “If the tax rate was 100% I should stop working.”

    Starve yourself to death would you? In protest?

    You mean that taxes are 100 %, but employee is given food if he works, not given food if he does not work?

    In that case the tax rate is not 100 %. They payment is in food, not money, but nevertheless the tax rate is not 100 %.

    But if everyone gets the same food, health care, etc, regardless of contribution to work, then tax rate is 100 %. And then everyone can just stop working if they don’t enjoy doing it for nothing – they still get the same living.

    Of course, you may then invent a system where everyone gets the same food regardless of working of not, but those who do not work as they are told to, will be shot. A bit like the Gulag, you know. Which seems to be the system that is favoured by those who insist that the Laffer curve does not exist.

  16. Dear Tim,
    Your instincts about Will Hutton are correct. The Mirrlees Review (not by Mirrlees) is one of the most sterile, static, tedious and practically irrelevant studies that I have ever forced myself to peruse. Even worse than the Meade Report, which was at least internally consistent. Like Meade, there is far too much emphasis on highly detailed fairness (static welfare economics) and far too little on the practical effects of a tax system on growth, let alone the lives and behaviour of real people.

  17. When comparing economies (econometric “proofs” of the Laffer Curve) the vertical axis is always scaled to GDP, otherwise there is no way to put multiple countries on one graph. On single economies, the graph is generally a dimensionless “for illustrative purposes only” because you can’t actually measure one economy at different tax points, because only one copy of it exists in the world. But, the scale doesn’t matter much anyway, not in fantasy economics.

    It’s interesting to see people are ducking and diving around a very simple and clear argument I’ve made. Whether or not communist economies are desirable or practical or would need Gulags is neither here nor there. The 100% case is at the extreme end of the graph. At lesser degrees of taxation, populations with different philosophies will be prepared to hand over lesser or greater quantities of their income for the provision of state services. What we’re interested in with the Laffer Curve is what happens to tax revenues under those different circumstances, all else being equal. Are they zero, or are they 100% of GDP? The answer is, as I’ve shown, the latter. You can take it down to, say, 80%, and then nudge up to 100 in paradox of the heap stylee, if you like.

    The key error seems to be ignoring the public sector. The presumption being made is that taxation just disappears. It doesn’t. Governments spend it. That is why they tax. The only way that you can get zero tax returns for 100% tax is to have the government tax every penny in the economy, once, then hide it in a vault. With no money left, revenues are indeed zero. But that’s not why governments tax. They tax to spend. In that case- in which revenues recirculate as spending, the 100% case is correct.

    This is a very simple point and argument. The basis of the Curve is that revenues are zero at each end, and higher in the middle. Therefore there must be a maximal point between the two. But there isn’t. QED and all that.

    There are certainly compelling arguments that moderate tax regimes will be more productive in goods and services. There are utterly compelling arguments for not being communists. But the Laffer Curve has a specific argument attached; that there is a “sweet spot” of maximal revenue for tax rate, above and below which tax revenues will be lower. This is not true. There is no Laffer Curve.

  18. Also, pjt-

    A bit like the Gulag, you know. Which seems to be the system that is favoured by those who insist that the Laffer curve does not exist.

    It’s fun to be accused of being a commie, considering I’m a rabid libertarian who wants zero taxes in libertopia. I just don’t like bad arguments, because they make us look stupid. The Laffer Curve is a very bad argument, and it makes us look very stupid indeed.

    Look, the gravest error is to try to win a leftie around by promising that right wing policies will produce left wing results. They won’t. You don’t argue that lowering taxes will raise taxes. It’s assinine. You cut taxes to starve the government, not to feed it.

  19. Isn’t there an empirical way to look for evidence of the Laffer Curve?

    Supposing one were to do a salary distribution histogram, showing the incidence of pre-tax income in increments of £1,000, ie £1-£1,000, £1,001-£2000, £2,001-£3,000 etc.

    I presume it would produce a skewed distribution, with a peak, I would guess, around £15k, and a ‘long tail’ to the right containing the Bob Diamonds and Wayne Rooneys of this world.

    Now, the 40% tax band kicks in, for a single person, at £42,475.

    If people do indeed shift their behaviour because of tax rates, wouldn’t one expect to see an anomalously high number of people in the £41,001-£42,000 range, and an anomalously low number of people in the range £43,001-£44,000?

    IE wouldn’t such a graph show whether people did indeed ease back a bit on overtime, sidelines, chasing promotion etc, because the next pre-tax £1k was worth £200 less than the last one?

    Naturally, the place to look for a kink in the graphs only applies to single people, because of their tax allowances. I presume it could be repeated for with different graphs for people with 1, 2, 3 children etc.

    I don’t claim to have the figures, and I don’t have time to do the exercise- but has anyone done this sort of thing already?

    (I certainly recall reading about a similar thing in France a few years ago- IIRC a EUR5,000 “solidarity tax” payable if one earned EUR65,000. Self-employed people approaching that point would take a bit more holiday and stay below the threshold. )

  20. CJ, the problem there is that you’re not looking at the claim the Laffer Curve makes. You’re studying peoples’ behaviour withing a particular (complex) tax system with different rates and so on. The LC is instead a claim about an “aggregate” tax rate and has nothing to say about whether peoples behaviour shifts due to hitting a particular tax band. It is a simple claim that there is a relationship of a particular type between aggregated tax revenue and the aggregated tax rate.

    There is no doubt that people will alter their behaviour to suit different personal economic circumstances, including taxation rates. Laffer is “macro”, not “micro”.

  21. IanB said (#20) “What we’re interested in with the Laffer Curve is what happens to tax revenues under those different circumstances, all else being equal. Are they zero, or are they 100% of GDP?”

    But isn’t the point of the Laffer curve that tax revenues go up or down, not that they go up or down as a percentage of GDP. Because the argument behind Laffer is that raising tax rates reduces GDP.

  22. CJ: I don’t think the data would be strong enough, in the absence of ridiculous non-tapered taxes producing marginal rates over 100% like the French one you cite (which would, obviously, have a significant impact).

    Most workers earning close to higher-rate are salaried rather than waged, and so overtime isn’t relevant; they’re in highly-enough paid jobs that slacking off and hoping not to get fired is a dangerous strategy; and an extra GBP500ish (after NI) of your next GBP1000 pay rise is still a nice thing to have.

    You’d need people to have far more direct control over their earnings than they do to detect the slight marginal impact of taking home 50ish% instead of 70ish% of a rise in salary.

Leave a Reply

Name and email are required. Your email address will not be published.