{"id":7582,"date":"2022-09-05T20:00:00","date_gmt":"2022-09-05T19:00:00","guid":{"rendered":"https:\/\/www.taxpolicy.org.uk\/?p=7582"},"modified":"2023-02-07T11:47:08","modified_gmt":"2023-02-07T11:47:08","slug":"designingwindfalls","status":"publish","type":"post","link":"https:\/\/heacham.neidles.com\/2022\/09\/05\/designingwindfalls\/","title":{"rendered":"How to design a \u00a330bn windfall tax on the energy sector"},"content":{"rendered":"\n
If we wanted to raise significantly more than the Government’s \u00a35bn Energy Profits Levy<\/a>, how would we do it? <\/strong><\/p>\n\n\n\n This is a further explanation and expansion of points I make in Panorama’s programme on the energy crisis<\/a>, broadcast on 5 September, and follows my previous post discussing two serious flaws<\/a> in the Energy Profits Levy – the relatively modest “windfall tax” the Government introduced in May.<\/p>\n\n\n I’m not going to get into the rights<\/a>\/wrongs<\/a> of windfall taxes, other than to note that the political case for a more ambitious tax may be hard to resist if energy companies make \u00a3170bn of “excess profits”<\/a> over the next two years, when at the same time it seems increasingly likely that the Government is going to spend \u00a318bn<\/a> or even \u00a329bn<\/a> keeping energy prices down for consumers. <\/p>\n\n\n The big problem is: how do we define an “excess profit”? The apparent<\/a> Treasury leak<\/a> of the \u00a3170bn figure didn’t bother to say1<\/a><\/sup>Which is one reason to be suspicious of the leak, and of the figure<\/span>.<\/p>\n\n\n\n It was relatively easy for the last two big UK windfall taxes.<\/p>\n\n\n\n Note the important common features: one-off taxes, applying to clearly identified windfalls, and doing so retrospectively. These are good design features which we should seek to copy. The retrospective element is counter-intuitive – normally we’d say retrospective taxation is undesirable, even wrong, but in this case retrospection is essential both to enable the windfall to be clearly identified, and prevent avoidance\/distortions. <\/p>\n\n\n\n The difficult question in this case is: how do we identify the windfall?<\/p>\n\n\n Here is the heart of the windfall, the European gas futures price:2<\/a><\/sup>Thanks to R for their help with this – the chart is from here<\/a><\/span><\/p>\n\n\n\n It’s this that feeds into electricity prices generally, because the marginal price of electricity is set by gas. There is an excellent explanation of this here<\/a>.<\/p>\n\n\n\n In principle, we could calculate what energy company profits would have been in a parallel universe where energy prices hadn’t changed since January 2022. Then deduct that from the actual profits, and tax the difference. Fantastic, but for the minor detail that this is impossible – no complex business has systems\/accounts that enable a calculation of this type to be made (i.e. because energy prices don’t go straight to the bottom line as revenue – energy prices are also a cost for most energy companies, and both costs and revenues are often hedged, energy is often sold on the basis of fixed prices, etc etc etc). If you ignore these complexities and just tax the immediate revenue impact of the increase in prices, then you end up with the Spanish windfall tax<\/a>, which was something of a disaster<\/a>. <\/p>\n\n\n\n That suggests we need to look at profits, not prices, and apply some kind of revenue trend<\/a> approach, for example saying that 2021 is the baseline profit, and any increase in 2022 is excess profit which we’ll tax. But that’s extremely random, and really just punishes companies who had a poor 2021. If Shell and BP go on to make identical profits in 2022, it can’t be right that we tax BP more, just because in 2021 Shell’s profit was higher<\/a>. <\/p>\n\n\n\n Alternatively, we could follow the approach adopted by the various wartime excess profit taxes<\/a> which looked at the average profit during a prior period. This was proposed<\/a> as a way of taxing excess profits resulting from the pandemic, but to my mind is a bad fit for the energy sector, given the cyclical nature of energy company profits (exacerbated by their very poor performance in the 2020 pandemic):<\/p>\n\n\n\n I wouldn’t altogether discount this approach, particularly if we look across a very long period (say 20 years), but even then the result would be highly contingent on the precise period chosen (and where it cuts across the economic cycle), and therefore somewhat arbitrary. The extensive caselaw <\/a>on the UK’s wartime excess profits duty (which shaped a surprising amount of modern tax law) is testament to quite how arbitrary and uncertain such taxes can be.<\/p>\n\n\n\n Other more sophisticated excess profit tax approaches have been proposed, e.g. categorising individual sources of profit<\/a>, but they don’t seem very applicable to the problem we’re trying to solve.<\/p>\n\n\n Most<\/a> of the large profits from the current crisis are being made elsewhere in the world – particularly in Gazprom, Saudi Aramco<\/a>, and a plethora of US energy companies. Some people (okay, the Lib Dems<\/a>) have suggested we tax those companies’ UK subsidiaries. <\/p>\n\n\n\n The difficulty with this is that in most cases all, or almost all, the profit is made outside the UK – so we have nothing to tax. Gazprom is an interesting exception – it has a UK trading arm<\/a> with a reasonably significant market presence – and that’s probably why the Lib Dems name-check it specifically. The minor problem is that the company isn’t doing so well<\/a> these days3<\/a><\/sup>And its profits were a fairly modest \u00a3300m even before it started running into trouble<\/a><\/span>.<\/p>\n\n\n Another suggestion has been to tax listed energy companies on their capital appreciation. This has a number of difficult legal issues, and in particular potential conflicts with tax treaties and EU law. I discuss this further here<\/a>.<\/p>\n\n\n So the best solution may be the simple one of imposing a special top-up tax on the profits of energy companies from 26 February 2022 to whenever it is that market prices settle down. I’d tentatively set out the following design features:<\/p>\n\n\n\n If the actual windfall profit ends up being around the supposed leaked Treasury figures, then it’s not unrealistic to think such a tax could potentially raise \u00a330bn over two years.<\/p>\n\n\n\n However, it’s important to stress that we shouldn’t start taxing a windfall before a windfall has actually been made. Right now, BP and Shell had a very good Q2, and a decent Q1 (ignoring BP’s huge loss from the forced sale of its Russian business8<\/a><\/sup>I would allow such losses to be deductible in the proposed windfall tax, as they’re part of the IAS consolidated accounting profit – but I can see a case for excluding them, given that companies could control the timing of disposals to avoid tax. One option would be to tax\/relief gains\/losses from 26 Februrary 2022 up to the date the tax is announced, but not afterwards<\/span>). Let’s see what happens next. When it comes to windfall taxes, there’s no time like the past.<\/p>\n\n\n\nThe politics<\/h2>\n\n\n
The design challenges<\/h2>\n\n\n
\n
Three sophisticated approaches that won’t work<\/h2>\n\n\n
<\/figure>\n\n\n\n
<\/figure>\n\n\n\n
A lovely idea that can’t work<\/h2>\n\n\n
More radical ideas that are legally complex<\/h2>\n\n\n
A simple approach that should work<\/h2>\n\n\n
\n
\n\n\n\n