The reform of England’s GCSE grading system, which will be big news in the coming weeks as the first set of exams to use it produce results, is afflicted by two competing policy objectives. They cannot both be satisfied.

This incoherence, a product of a complicated, characteristically over-politicised and often-redrafted set of reforms set in train five years ago by Michael Gove, explains a lot of the confusion around the change.

The first policy objective is made clear in a letter from Mr Gove to Glenys Stacey, then chief executive of the exams regulator Ofqual, in February 2013.

Ofqual had just rejected Mr Gove’s plan to turn GCSEs into a now-largely-forgotten new qualification, called English Baccalaureate Certificates (EBCs). This plan itself had replaced a plan by Mr Gove to reintroduce O levels.

Ms Stacey had advised that moving to EBCs was just too complicated and ambitious. The former education secretary reluctantly accepted this, so the GCSE name was retained. But he wanted to press on with plans to toughen up GCSEs with “harder” content.

In his letter to Ms Stacey, Mr Gove said: “I consider there to be a strong case for the reformed GCSEs to have a new grading scale, to reflect the step change in expectations for pupils.”

And we do indeed have this new system. So the first policy objective was to introduce a new grading scale to reflect the fact that standards will have been raised: expectation regarding the quality of the work from pupils will have increased.

But the regulator Ofqual, perfectly reasonably, given its concern not to disadvantage one year group of pupils in comparison with another, has advised that grades from the old system should be comparable to grades from the new.

That is: the standard of work required to get, say, a grade B in the old system will be the same as that required to gain a 6 in the new one. So, as a competing policy objective, the new grading system seeks to maintain standards from previous years.

It is true that the content of exam syllabuses has changed significantly in this set of reforms, with Mr Gove introducing more subject matter and scrapping coursework, in a move widely seen as raising demands on many pupils.

Yet Ofqual’s grading regime has been set up to ensure that any increase in difficulty is not reflected in a reduction in the proportion of pupils gaining good grades. So, if pupils are found to have struggled, on average, with these new courses, then grade boundaries will need to be relatively low to ensure the proportion of pupils getting good results stays broadly constant.

Confusing everyone

So what, then, was the point of changing the grading scale?

It seems hard to find an answer from the paper trail. With each old GCSE grade now finding an equivalent in the new numerical system, all the new structure does is, first, introduce a new grade of 9 above the old A*; second, instigate two sub-grades at the equivalent of the old C; and, third, contract the number of grades at the bottom of the scale from three (E, F and G) to two (2 and 1).

But if greater differentiation were needed at the top end (and I go along with this view that it is likely to needlessly add to some pupils’ anxiety levels), then a simple renaming of A/A* grades to, say A1, A2 and A3 would have sufficed. It is not clear what reducing the number of lower grades achieves.

So this only leaves us with the reformed C grade. This has been the subject of much confusion, with the exam boards and Ofqual stating that pupils achieving a new grade 4 will be seen to have achieved an old grade C, or “passed”. This is in line with its objective of having a new “pass” grade which is equivalent with the old one.

Meanwhile, the government wants to publish schools’ success rates both at this level and in the slightly more demanding grade 5, which it deems a “strong pass” .

Politics seem to me to be behind the move. Ofqual’s 2014 consultation on the new grading system made it clear that ministers wanted a new C-grade benchmark to reflect their desire to raise statistical expectations of pupils through the new exams.

Ofqual was asked to make them more demanding, in line with the performance of pupils in countries which do well in international tests. The new grade 5, which the 2014 Ofqual document said would be up to “two thirds of a grade higher than that required for a current grade C”, is the result.

So that’s it: this whole rigmarole seems to come down to giving ministers the ability to specify a slightly increased set of performance metrics in reporting pupil and school results.

Some will respond that the increased demand of the exam syllabi themselves is rightly being recognised through new grades. But I would argue that the fact that new grades are being made to correspond with the old renders that argument worthless: if a signal was to have been sent about radically “improved” qualification content, it would have been better to have renamed the exams completely.

But this brings us full circle, with the new grading system being suggested after Ofqual had advised against changing the name of GCSEs.

Education policy often verges on the surreal. We must hope that this latest confusing scenario has not cost pupils, teachers and the country more than the £500,000 already reported as having been spent communicating it.

Warwick Mansell is a freelance education journalist and author of Education by Numbers. You can read his back catalogue here. He tweets @warwickmansell

Want to keep up with the latest education news and opinion? Follow Tes on Twitter and like Tes on Facebook.