Written evidence submitted by Tim Worstall (Senior Fellow at the Adam Smith Institute)


This is submitted on a personal basis.


Wealth taxation is a currently popular idea in certain circles. There are two main problems with the implementation of such a plan – it won't raise much revenue and standard economics is rather against the idea. The justification that is proffered to override those two points is that wealth inequality is so great these days that its reduction is the grand benefit to be gained. The problem here being that we don't know – have no clue about – what the current wealth distribution is.


The obvious action to be taken is therefore that we find out what that wealth distribution is so that we can make informed decisions about whether we wish to reduce the inequality of it or not. Fortunately this is not difficult and the Office for National Statistics (ONS) most certainly has the technical ability to do the work. All that would be required is a little guidance as to how it should be done and the United Kingdom could be the first country in the world to measure this part of the economy correctly. The specific method to be used comes directly from the currently fashionable works of Mssrs Piketty, Saez and Zucman.


Revenue from wealth taxation


There's not all that much tax revenue to be gained from wealth taxation in the UK – not over and above the amount that we already collect. As the ONS points out total household wealth (in 2018)(1) was £14.6 trillion. A substantial sum that we should be able to lift some revenue from. As an aside, yes, all things are, ultimately, owned by households. There is no other, outside that already owned by government, pot of wealth out there. There is a problem with trying to tax this sum though:


Property Wealth (net)                            5.09             

Financial Wealth (net)              2.12             

Physical Wealth                             1.25                           

Private Pension Wealth          6.10             

Total Wealth                                      14.63



The idea that we're going to start taxing pensions wealth after decades of tax incentivising people to save for their old age is more than a little odd. Taxing property wealth is clearly possible because we already do this more than any other rich nation. As the OECD points out (2):


Across the OECD, immovable property taxation made up around 3% of the total tax take and a little over 1% of GDP in 2013. Countries with a high property tax-to-GDP ratio include the United Kingdom, Canada and the United States, while it is almost nil in countries such as Luxemburg, Switzerland and Mexico.






Physical wealth such as the antique thimbles collection will be difficult to tax so that leaves us just with financial wealth as a possible target. We know who owns the shares and bonds, where they live, taxing them is possible.


However, if we try for a substantial levy, tens of percentage points of the total in order gain reasonable sums, everyone has to be selling their stock at the same time in order to pay the tax bill. Something that would rather destroy the value of the very thing being taxed, such a wave of portfolio liquidations. If we instead tax at a rate that can be paid from the income of such assets, say 1 or 2%, then we're not going to gain all that much revenue. Further, we already do tax such income through the income tax system so it's not obvious that a great deal will be gained by it.


As a significant revenue raiser wealth taxation doesn't seem to offer much.


The idea of wealth taxation


The standard economics of taxation is against the taxation of wealth. Not for conservative (or even Conservative) reasons but because it's seen as counterproductive. Wealth is the product of investment and that's what makes the future richer. People delaying consumption in order to build things to be used in the future is regarded as a good idea. Why would we want to tax those making our children richer? After all, we get less of whatever it is that we tax.


Of course more detail is available in optimal taxation theory or the Mirrlees Review (3).


It's possible to reach the same destination through a different logical route. Every tax has a “deadweight cost”. This is the economic activity that doesn't take place simply because we have taxed at all. Yes, it's entirely possible that what the money is then spent upon makes it worthwhile. It's also true that at times we wish to tax something in order to reduce the amount of it – smoking, say, or a congestion charge. But all taxes and every tax has that deadweight cost. Which varies according to which tax we're imposing upon what.


We can list taxes by their deadweights. The lowest is “repeated taxation of immovable property” or, as it might be better known, the land value taxation of Henry George. Moving up to more economic activity expunged for a unit of revenue collected we have consumption taxes like VAT. Then income taxation, then corporate, then taxes upon capital and wealth until we top out with transactions taxes like the financial transactions tax (or “Robin Hood Tax”). Viewing our tax system as a method of gaining revenue – before we start to ponder other changes we'd like to institute through taxation – we want to be at the beginning of that series, not the end. Raising revenue by taxing wealth and or capital costs us more than gaining the same government income through income or land taxation. We therefore don't want to do it unless we've some other very good reason other than simply revenue.  


The current reason – wealth inequality


A current insistence is that wealth inequality is so great that we would all benefit from its reduction. A more equal society, as with Wilkinson and Pickett in The Spirit Level, would be something that aids the entire society. Leave aside that contention as a political, not economic, matter and consider the following. We don't know what the wealth distribution is. We've not a clue. We know what it is before the impact of all the things government does but we have no knowledge, because no one measures it, of what it is after all of those things. This is problematic for the policy relevant, what we need for decision making, information is how much inequality left unequalised is there? Not the current query of what would it be like if nothing were being done?


This is made entirely clear in the usual source document for proof of that wealth inequality, Saez and Zucman (4). To quote at length:


Our definition of wealth includes all pension wealth—whether held on individual retirement accounts, or through pension funds and life insurance companies—with the exception of Social Security and unfunded defined benefit pensions. Although Social Security matters for saving decisions, the same is true for all promises of future government transfers. Including Social Security in wealth would thus call for including the present value of future Medicare benefits, future government education spending for one’s children, etc., net of future taxes. It is not clear where to stop, and such computations are inherently fragile because of the lack of observable market prices for this type of assets. Unfunded defined benefit pensions are promises of future payments which are not backed by actual wealth. The vast majority (94% in 2013) of unfunded pension entitlements are for Federal, State and local government employees, thus are conceptually similar to promises of future government transfers, and just like those are better excluded from wealth.


Everything government does which changes the current wealth distribution is specifically excluded from our estimation of the wealth distribution. This is obviously not a useful measure of that distribution for policy making purposes. Our task is to try to decide whether more needs to be done to change that distribution. An observation of what it would be without what is already done isn't the right starting point for that deliberation. This has been called “Worstall's Fallacy” (not, I hasten to add, originally called that by me but so named because I make the point so often): To propose what must be done before considering what is already done to solve the problem.


There is another point here of course. If we do not count tax and redistribution in our estimates of the wealth distribution then we cannot change the wealth distribution through tax and redistribution. The effects of what we do will not appear in our calculation of what remains to be done. Again, this is not helpful as information flowing into our decision making.


We have largely solved this problem when talking about incomes, as we have not done with wealth. ONS regularly calculates the “Gini” (a measure of inequality) for UK incomes and presents us with the original numbers from pure market incomes and then adjusted for the effects of taxes and benefits. An example (5):




It is that disposable income Gini, the lowest one and accounting for taxes and benefits, which is the policy relevant number used in British political discussions. As it must be of course even for those who say we must be more like Sweden. For the original income inequality isn't all that different between the two countries, Sweden does more redistribution. Thus the use of the number which shows how much more we've got to do to be more like Sweden with their Gini of perhaps 0.25 – all achieved through that tax and redistribution.


Some places don't account for everything that is done though. The American Official Poverty Measure doesn't include the effects of, as one example, the EITC when calculating the number below their poverty line. The EITC being roughly the equivalent of our Working Tax Credits. Thus in the US system it's possible to argue in favour of doubling that EITC on the grounds of reducing poverty but by the system of measurement being used reduce poverty by not one whit. We don't make that error with incomes – possibly because those who instituted the WTC wanted to be able to claim a reduction in poverty – but everyone, everywhere, still makes it with respect to wealth.


We do not take account of what is already done to modify the wealth distribution when measuring the wealth distribution to decide upon what should be done. And, clearly, we should.


The method to use is there in Saez and Zucman. It's the method they use to calculate their wealth figures anyway. Capitalise the income streams that stem from an investment. This is not a complex matter and can be done, should be done, to the effects of taxes and redistribution, to the welfare state itself. Only once we've done that can we know what the current wealth distribution is – an essential prerequisite of any plans to change it.


The recommendation


To instruct the Office for National Statistics to prepare estimates of the wealth distribution on the same bases used for the income distribution. To calculate both the market, or original, distribution and also the one after the impact of taxes, benefits and government spending upon services.




1)                  https://www.ons.gov.uk/peoplepopulationandcommunity/personalandhouseholdfinances/incomeandwealth/bulletins/totalwealthingreatbritain/april2016tomarch2018


2)                  https://www.oecd-ilibrary.org/sites/9789264254053-5-en/index.html?itemId=/content/component/9789264254053-5-en#:

3)                  https://www.ifs.org.uk/publications/mirrleesreview

4)                  https://gabriel-zucman.eu/files/SaezZucman2014.pdf

5)                  https://www.ons.gov.uk/peoplepopulationandcommunity/personalandhouseholdfinances/incomeandwealth/bulletins/householdincomeinequalityfinancial/financialyearending2020provisional#



November 2020