To tax, or not to tax. that isn’t actually the question.

A £22 million estimate has spent this week as an international news story. What that tells you is not really about tax. It is about how facts behave in a charged debate, and it matters to anyone deciding where to put their people, or indeed where people decide to put their money.

Since the weekend, one piece of tax analysis has travelled a very long way. A single estimate, that Scotland's 48p top rate of income tax may have collected around £22 million less in its first year than the rate it replaced, has been carried by the Sunday Times and City AM, seized on by broadcasters like Andrew Neil, mused over by the respected Adam Smith Institute, written up approvingly as far afield as Washington, and turned into a stick to beat the Scottish Government with. It has also been dismissed, by serious and numerate people, as noise, and by others as nothing more than predictable political point-scoring from those with little time or love for the SNP-led administration.

Both reactions tell you more about politics than about tax. So let me do the unfashionable thing and look at the numbers instead. I spent two years doing just that.

What I learned asking the questions

For part of my time at Holyrood I was deputy convener of the Public Audit Committee. We had HMRC, Audit Scotland and senior government officials in front of us, more than once, on precisely this subject.

What stayed with me was not a clean answer. It was how carefully everyone at the table avoided giving one. These are the auditors and the tax authority, not people prone to drama, and their honest position, session after session, was a version of the same sentence: we believe it raises money, we cannot yet tell you how much, and the top of the scale is the part we are least sure about.

That is not a scandal. It is the actual state of the evidence, and this week it has been buried under a great deal of noise from people who are very sure indeed.

Raising a rate is not the same as raising the money

Here is the part the shouting skips over, and it is not political. It is arithmetic and human behaviour.

At most income levels, raise the rate and you raise the money. People cannot do much about it. At the very top, they can. A higher earner can put more into a pension, take company dividends instead of salary, defer income, or occasionally move south. None of this is exotic. It is what the evidence, from Scotland, the rest of the UK and abroad, says reliably happens.

The structural numbers show it. If the Scottish tax base held exactly as it is, the country's higher rates would raise around £1.8 billion more than the rest of the UK's. The actual net gain is closer to £1 billion. The gap, which the Scottish Fiscal Commission calls the "tax base performance gap", is behaviour and slower earnings growth doing their quiet work. On the top rate alone, the Commission's own modelling has around 85% of the roughly £53 million it might raise disappearing into that behaviour, leaving perhaps £8 million.

Then again, the theory still stands: higher taxes pay for "things". Nothing is free, as we are oft reminded.

So the direction of this week's story is not mad. It is the certainty around it that is.

The fiction on both sides

Here is the detail that should give every headline writer pause.

The analyst who started this is no partisan outsider. He is a tax expert who until recently sat on the Scottish Government's own advisory group, and he was careful to say that the 48p rate was always a political symbol rather than a serious revenue measure, and that £22 million is a rounding error against a Scottish income tax take of around £19 billion. His point was never that Scotland had blown a fortune. It was that the symbol may quietly cost a little rather than raise it.

The serious critics make the mirror point. £22 million is roughly 0.12% of the total, comfortably inside normal forecast error, timing effects and the classification noise you expect in early HMRC data. On that reading, the honest answer is that we cannot yet say the rate has lost anything at all.

Notice what has happened. A tiny and deeply uncertain figure has been claimed as proof by one side and waved away as nothing by the other, and both are overreaching. In a debate this charged, the facts matter more than the fiction, and there is fiction flying in both directions this week.

Why a boardroom in England should care

This is where it stopped being a Scottish story. This is where it became less about how much senior execs in EH1 postcodes pay in tax and more about the state of the UK economy, the role of devolution and sort of news stories which affect decision making in boardrooms in EC1.

If you employ senior people, the divergence is a live cost. Keeping a £125,000 hire whole costs meaningfully more in Edinburgh than in Manchester or London, and someone in your finance team already knows the figure. Firms based in England, and overseas companies choosing where to land in the UK, increasingly factor the differential into where they place teams, where they grow, and occasionally whether they stay. Of course, you can live more cheaply north of the border than in London, and many choose to do so. Spoiler: I did just that.

The timing of this furore is not incidental. With a new government in Westminster weighing its own tax choices, Scotland's experience is being read in the Treasury as a live case study in what happens at the top of the scale. Whatever London decides, the size of the gap between Scottish and UK rates could move again, and it is the gap, not the headline rate, that drives the behaviour.

The honest answer, which I trust you came here for, is that we simply will not know for some time. Once all that tax intake has undergone a good scrub in the HMRC wash-house, been fed into the fiscal framework, and spat out the other side of the block grant adjustments, then, and only then, will a future Public Audit Committee in Holyrood be able to write a report on it all which will generate more headlines.

So the question in the headline really, to tax or not to tax, is the wrong one. The useful questions are what a tax actually raises once people respond, and who quietly decides to live, hire or grow somewhere else because of it, and how a simple figure from a respected analyst gathers political momentum to make front page headlines in hours. Answer those honestly and you are ahead of most of the debate, and most of the boardroom.

The lesson of it all? £22 million reasons to be careful what you say, how you re-say it, and where to look for the elusive truth amidst it all.

Jamie Greene served as deputy convener of the Scottish Parliament's Public Audit Committee, where he questioned HMRC, Audit Scotland and senior officials on Scotland's finances. He founded Kempock, which helps organisations understand how devolved policy affects real decisions.