This article, or something very like it, was in the Scottish Sunday Herald last weekend:

When I was giving evidence to the Finance and Constitution Committee at Holyrood earlier this year I said that I had never come across a country more interested in national income accounting than Scotland. The debate on GERS - the decidedly oddly named Government Expenditure and Revenue Scotland statement - proves this. The latest version was published this week.

My argument on GERS is that the data it provides is what I have technically described as ‘crap', which, as I also explained to the parliament, is a term that I use when teaching on economic data quality to students to describe ‘completely rubbish approximations'. It's my contention that this is what Scotland has had to deal with for many years. This needs to be explained.

GERS has two sides to it, as its name implies. On income until 2016 twenty-five of the twenty six figures that made up the data were estimated. I am pleased to say that this ratio has improved slightly this year because better, Scottish sourced, data is now available on some locally devolved taxes. However, let's not get too carried away: they make up a quite small part of total income, and even the team preparing GERS are candid enough to admit they cannot state with any confidence the degree of accuracy to which over a third of the income is stated. This includes some pretty significant figure, like corporation tax income, whilst oil estimates, which have been revised this year, still look open to question as well. The reality is that not nearly enough tax data is collected in Scotland as yet to be confident about many of the figures in GERS. I actually go further in my criticism than that though: I have suggested that this failure to collect data is deliberate. Politicians in Westminster are likely, in my opinion, to want the wriggle room that this creates for them.

Many argue that the expenditure figures in GERS are more reliable than the income data because a significant part of that spend is devolved to Scottish government control. This, of course, is true. But that's not the whole story there either. The parameters in which the spending arises are set in London in many cases, and they too provide the money, setting out what Scotland can pretty much do as a result.

And then in very many parts of GERS Scotland is simply attributed with a part of total UK spending. In many cases, however, such as defence, foreign affairs, and quite possibly many other policy areas where this attribution arises, Scottish public pinion clearly indicates that if it had the chance Scotland would make very different spending decisions to those for which it is currently charged. GERS, however, does not allow for that: as the Fraser of Allender Institute have acknowledged, GERS assumes Scotland is a mini-part of the U.K. and no compensation for its higher levels of spending in some areas is reflected in other costs apportioned from the rest of the U.K., meaning that the supposed Scottish deficit may be seriously overstated as a result. It is also possible that the tax revenue generated by the spend outside Scotland deemed in GERS to be for Scottish benefit should also be, but is not, credited to the GERS revenue account. If that's the case then there is a serious accounting flaw in the whole GERS process that undermines all the data it supplies.

What is the net consequence of all this? It's that in my opinion three things. The first is that a lot of time and effort is now spent discussing a report that's full of cobbled together data, however will meaning the statisticians who prepare it might be. Second, that statement suggests, on pretty unsound economic and accounting foundations, that Scotland remains dependent on the UK when that might not be true and when Scotland is, according to regional trade data, the only part of the U.K. that is currently running a trade surplus with the rest of the world. But perhaps most important of all, the GERS data leaves the Scottish Parliament completely in the dark in the real impact of the decisions they have taken and have to take.

My suggestion is that this is deliberate. Scotland has been given want look like devolved powers but has little relevant, reliable or usable data on which to decide what to do with them. That's a recipe for hit and miss government and if in occasion that's what Scotland has has got that's firstly what Westminster wants and, second, Holyrood is not to blame.

But what really surprises me is that we do not hear Scottish ministers making a fuss about this. Nor do I see them queuing up to demand the better quality data Scotland could and should have on which to make decisions if only the political will to create it existed. That is a mistake by all Scottish politicians that they alone can correct.

But what they can't do is put GERS right. That's because it hardly recognises Scotland's right to decide on almost any issue of consequence. As a result what is really needed now is a whole new system of national accounting for Scotland, whether devolved or independent, that is in charge of its own decision-making. Discussing GERS in detail is, then, a waste of time because it's not designed to be meaningful, and isn't, in my opinion. And that's the issue to be addressed.

PS: I am giving evidence in Holyrood again on 19 September