A think tank, EDSK, published a paper<\/a> today looking at this point. Unfortunately, they have not actually undertaken their own analysis, just made some simple calculations based upon previously published figures. The problem is that those figures are, in our view are highly questionable, and perhaps useless. Hence this is a case of “garbage in, garbage out”, and we would caution against taking any figures in the report seriously.<\/p>\n\n\n\n For the same reason, we would caution against taking any<\/strong> of the various figures floating around seriously, unless and until a full analysis is undertaken.<\/p>\n\n\n\n Credit to EDSK – their paper readily admits its own flaws, right at the end – key effects were not taken into account. But these are significant issues which can’t just be ignored:<\/p>\n\n\n\n Any estimate of the revenue impact rests upon a key question: what percentage of children would leave private schools if VAT was imposed? The paper uses two previous estimates: 5% and 25%. It treats them as higher and lower upper bound estimates:<\/p>\n\n\n\n But the 5% and 25% figures shouldn’t be treated so seriously.<\/p>\n\n\n\n The 5% figure comes from the 2019 Labour Party manifesto. It took a price elasticity estimate from an IFS analysis<\/a> which looked at the impact of changes in school fees on the rate of children going to private schools. Then the Labour Party simply applied the elasticity to a 20% price increase.<\/p>\n\n\n\n This was not a good approach. The IFS paper looks like a serious piece of work, but it looked at much smaller prices than 20%, and occurring over a long period. So it is likely incorrect to assume the IFS elasticity holds for an immediate 20% VAT increase, and this error surely means that the Labour Party figure was understated. On the other hand, the IFS found price sensitivity only at entry points – ages 7, 11, 13, and so it’s not correct to simply apply this elasticity to all private school pupils. That would tend to over-state the effect. Taken together, these effects render the 5% estimate of limited and perhaps no use.<\/p>\n\n\n\n The 25% figure comes from a slightly mysterious survey <\/a>of heads and parents at 21 schools by a consultancy engaged by the Independent Schools Council. The mystery being that details of the methodology are scant and, even in principle, surveying parents and headteachers in a mere 21 schools seems unlikely to reveal much about what would actually happen across the country if school fees increased. The likelihood of conscious and unconscious bias is obvious:<\/p>\n\n\n\n Exactly how the calculations were carried out is not revealed in the paper, but there is no evidence of any statistical analysis, and results are presented to two decimal points without any discussion of statistical error (or indeed even a single mention of any statistical tools).<\/p>\n\n\n\n Hence we would regard the 25% figure as meaningless. We don’t think EDSK should have taken them as upper\/lower bound estimates, or even used them at all.<\/p>\n\n\n Having spoken to a variety of economic and education policy experts, we believe a proper analysis would look something like this:<\/p>\n\n\n\n So the figures in this paper are just guesses multiplied by guesses. The difficult and uncertainty analysis that is omitted is where the truth actually lies. To be fair, the paper really admits that at the end.<\/p>\n\n\n Something the report gets right is that the cost for a private school of imposing 20% VAT would not be 20%. <\/p>\n\n\n\n At present, private schools suffer VAT on their expenses (\u201cinput VAT\u201d) but, unlike a normal VATable business, they recover little or none of this. <\/p>\n\n\n\n In other words, a normal business buys a \u00a31,000 computer. They pay \u00a31,000 plus \u00a3200 VAT, but can recover the VAT. Net cost: \u00a31,000. But for a private school, the net cost would be \u00a31,200 (or almost \u00a31,200). <\/p>\n\n\n\n Once school fees become VATable then the private school would be treated in the same way as a normal business. That computer would cost \u00a31,000 net. <\/p>\n\n\n\n The extent of this effect depends on the proportion of a school\u2019s costs that are currently VATable. The majority of the costs will be staff wages, and there\u2019s no VAT on that. The Independent Schools Association estimated the net impact would be 15% – we haven\u2019t seen underlying data supporting this, and so the figure should be used with caution. But it feels in the right ballpark. <\/p>\n\n\n Here we are able to comment fairly definitively. The concerns raised are weak and in some cases incorrect.<\/p>\n\n\n\n This section suggests that there is doubt as to how “closely related” supplies such as boarding accommodation will be taxed. <\/p>\n\n\n\n But that’s straightforwardly wrong. If education becomes VATable then closely related supplies will too. No further legislation would be required.<\/p>\n\n\n\n Then the report suggests there are obvious ways schools could escape the tax:<\/p>\n\n\n\n Trying to avoid the VAT by pushing boarding into a charity would be (naive) VAT avoidance, with no realistic prospect of success.<\/p>\n\n\n\n The paper mentions other potential complications, like the capital goods scheme and people paying years of fees up-front. But these are obvious points that I’d expect any legislation to deal with.<\/p>\n\n\n Technically it is straightforward to impose VAT on private schools. Things only get difficult if it’s done in such a way as to create arbitrary boundaries. For example, imposing VAT on schools charging (say) \u00a38k\/year or more would create a powerful incentive for a school currently charging (e.g.) \u00a310k to reduce its fees to \u00a37,999 and then pile on a series of individual charges for books, trips, etc.<\/p>\n\n\n\n Fortunately, politicians never<\/a> create<\/a> VAT rules<\/a> with arbitrary boundaries, so there is nothing to worry about here.<\/p>\n\n\n\n However at present we simply do not know what the revenue impact would be. We don’t know how schools would respond, and the extent to which the tax would be passed-on. We don’t know how parents would respond, and the extent to which pupils would leave the private sector. We don’t know how the State sector would absorb those pupils who would leave the private sector.<\/p>\n\n\n\n We’d therefore suggest disregarding the figures in the EDSK report, and the other figures sometimes referred to. At least until someone actually does the difficult job of undertaking a proper analysis.<\/p>\n\n\n\n More generally, when there\u2019s limited or bad data, the temptation is to use it anyway, because it\u2019s \u201call we have\u201d. That temptation should be resisted. Garbage in, garbage out. <\/p>\n\n\n\n <\/p>\n","protected":false},"excerpt":{"rendered":" We’ve been asked a few times to analyse the revenue impact of imposing 20% VAT on private schools. We’ve declined, because it’s a complicated piece of work which probably requires a large team with economic and educational expertise – the actual tax element is relatively small and straightforward. UPDATE 11 JULY: this article is now […]<\/p>\n","protected":false},"author":1,"featured_media":9914,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"ngg_post_thumbnail":0,"footnotes":""},"categories":[88],"tags":[171,152],"jetpack_featured_media_url":"https:\/\/heacham.neidles.com\/wp-content\/uploads\/2023\/06\/Screenshot-2023-06-07-at-09.43.43.png","_links":{"self":[{"href":"https:\/\/heacham.neidles.com\/wp-json\/wp\/v2\/posts\/9913"}],"collection":[{"href":"https:\/\/heacham.neidles.com\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/heacham.neidles.com\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/heacham.neidles.com\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/heacham.neidles.com\/wp-json\/wp\/v2\/comments?post=9913"}],"version-history":[{"count":14,"href":"https:\/\/heacham.neidles.com\/wp-json\/wp\/v2\/posts\/9913\/revisions"}],"predecessor-version":[{"id":10679,"href":"https:\/\/heacham.neidles.com\/wp-json\/wp\/v2\/posts\/9913\/revisions\/10679"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/heacham.neidles.com\/wp-json\/wp\/v2\/media\/9914"}],"wp:attachment":[{"href":"https:\/\/heacham.neidles.com\/wp-json\/wp\/v2\/media?parent=9913"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/heacham.neidles.com\/wp-json\/wp\/v2\/categories?post=9913"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/heacham.neidles.com\/wp-json\/wp\/v2\/tags?post=9913"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}![\"\"](\"https:\/\/heacham.neidles.com\/wp-content\/uploads\/2023\/06\/image-4-1024x403.png\")
<\/figure>\n\n\nThe questionable figures<\/h2>\n\n\n
![\"\"](\"https:\/\/heacham.neidles.com\/wp-content\/uploads\/2023\/06\/image-5-1024x303.png\")
<\/figure>\n\n\n\n![\"\"](\"https:\/\/heacham.neidles.com\/wp-content\/uploads\/2023\/06\/Screenshot-2023-06-07-at-10.05.48-1024x262.png\")
<\/figure>\n\n\n\n![\"\"](\"https:\/\/heacham.neidles.com\/wp-content\/uploads\/2023\/06\/image-7-1024x244.png\")
<\/figure>\n\n\n\n![\"\"](\"https:\/\/heacham.neidles.com\/wp-content\/uploads\/2023\/06\/image-8-1024x944.png\")
<\/figure>\n\n\n\nHow would an actual analysis be undertaken?<\/h2>\n\n\n
\n
![\"\"](\"https:\/\/heacham.neidles.com\/wp-content\/uploads\/2023\/06\/image-10-1024x277.png\")
<\/figure>\n\n\n\n\n
How much would the 20% VAT be offset by recovering VATable expenses?<\/h2>\n\n\n
What about the technical VAT concerns raised in the paper?<\/h2>\n\n\n
![\"\"](\"https:\/\/heacham.neidles.com\/wp-content\/uploads\/2023\/06\/image-11-1024x435.png\")
<\/figure>\n\n\n\n![\"\"](\"https:\/\/heacham.neidles.com\/wp-content\/uploads\/2023\/06\/image-12.png\")
<\/figure>\n\n\n\nOur conclusion<\/h2>\n\n\n