Last month, the likeable Chief Executive of HMRC, Jon Thompson, made waves when he revealed that a future ‘Maximum Facilitation’ customs arrangement with the EU could cost up to £20 billion per year. The largest proportion of this figure (roughly £13 billion) comes from the new cost to businesses of filling out customs declarations once we leave the Customs Union. The next £7 billion comes from additional customs compliance measures – primarily ‘rules-of-origin’ certificates. This figure was then dutifully recycled and publicised in the media with little scepticism or further enquiry. People and businesses recoiled in horror. On Tuesday, however, Mr Thompson was back in front of the Treasury Select Committee to clarify and caveat that initial figure. It has now become apparent that there are at least three reasons why it has almost certainly been hugely overstated – probably by at least half. Double Counting The first cause of this, and most straightforward, is that HMRC has double-counted the number of necessary customs declarations. In other words, the £20 billion figure includes the estimated cost of customs declarations for both UK businesses and EU businesses. This, apparently, was not made clear at the original evidence session as many media outlets implied – or at least failed to make clear – that this figure was not an accurate account of the cost solely to UK businesses. Indeed, once we discount the cost to EU businesses (something which is a problem for the EU, not us) then a whopping £6.5 billion is immediately struck off the total figure. Jon Thompson provided this much-needed clarity in Tuesday’s session, stating: “I think the thing which is possibly not as explicit as it should be… in terms of the evidence we gave you last time… is that that additional £6.5 billion occurs on the other side. So, £6.5 billion for UK businesses, and £6.5 billion for EU businesses… so the £13 billion, just to clarify the point… does not fall entirely on the UK side.” Straight off the bat, then, the total cost to UK businesses by HMRC’s own estimate is down to £11 billion to £14 billion rather than £17 billion to £20 billion. Getting your sums right The second cause for scepticism is the methodology that HMRC has used to calculate its £20 billion figure. As Jon Thompson made clear, there are two methodologies that could be used to estimate the cost of customs compliance in the ‘MaxFac’ arrangement. The first is a ‘top-down’ approach in which the cost of customs compliance is estimated as a percentage of the value of trade. The second is a ‘bottom-up’ approach in which one tries to estimate the number of expected customs declarations and then multiplies that by the estimated cost of each declaration. This is the methodology adopted by HMRC. As such, they have estimated that 205 million extra customs declarations will be required at an estimated weighted average cost of £32.50. This equals a cost of approximately £6.5 billion. However, if one adopts the alternative ‘top-down’ methodology, the final cost is significantly less. HMRC’s figure of £20 billion assumes that the cost of customs compliance of intra-EU trade in the future will be over 5% of trade value. This is far higher than the cost of customs compliance in other instances. As Jon Thompson has acknowledged himself, the cost of trade compliance for UK trade with the rest of the world is currently only between 1% and 3% of trade value. In addition, costs at the lower end of the spectrum (i.e. 1%) have been estimated in separate reports by both the World Bank and KPMG (Impact of Non-tariff Barriers as a Result of Brexit). In other words, HMRC’s methodology inflates the cost of customs compliance by anywhere between £11 billion and £16 billion relative to the alternative methodology. Changing behaviour The final reason to believe that HMRC’s figure of £20 billion is an over-estimate is because – as stated by Jon Thompson on Tuesday – it “is a static estimate on the basis of current patterns of trade and trade value”. In other words, the estimate makes the unlikely assumption that businesses will not alter their behaviour under new customs arrangements in an effort to reduce costs. One of the reasons that HMRC’s figure is so high is due to the observation that – at present – trade between the UK and the EU consists of a high volume of low value consignments. This means that the relative cost of customs compliance for each consignment will be higher. At present, because UK businesses trading with the EU do not undergo customs administration, there is little incentive to consolidate traded goods into larger consignments to reduce cost. However, once we leave, this incentive will be obvious. This is, after all, what we already do with non-EU trade. In fact, the value of UK export consignments to non-EU countries is a massive 16 times larger than that of exports to the EU. This fact was conceded by Jon Thompson himself when he told me that “it is of course possible that businesses might change their behaviour; they might package up consignments – they may have larger consignments and so on and so forth”. However, although this contingency is acknowledged by HMRC, it is not factored into their figures. It is difficult to predict the extent to which future business behaviour in this regard will reduce costs, but it is hard to believe that the aggregate savings would not run into the billions. As yet, HMRC have not published any detailed analysis on the costs of the ‘MaxFac’ model. This is regrettable as one would have hoped such a report to be made available prior to the announcement of total-cost figures. Given the significance of this issue, HMRC really ought to produce an in-depth report in which assumptions are not just presented, but justified. This report should make clear all the caveats and variables that are built into their calculations and respond to the serious criticisms of their figures that have been levelled by eminent thinkers such as Professor Graham Gudgin and John Mills. The ‘MaxFac’ model is the UK’s best hope of a frictionless trade arrangement with the EU after the implementation period – and it should not be dismissed on the basis of what is clearly only one way of slicing the figures.