Following the Post Office money?

At last the Post Office Horizon Inquiry touched on the operation of the Suspense Account.

Didn't find your answer?

Rodric Williams (former litigation lawyer at Post Office) clarified that the Suspense Account contained other transactions than those of the subpostmasters repaid “shortfalls” – those being associated with third party clients.

I think his intention was to limit the impact of release of sums to profit and loss (rather than those funds being applied to the subpostmasters’ accounts). I believe Post Office Limited use SAP as their accounting system. It does of course beg the question of why third party receipts and payments were allocated here too. Obviously allocating these bank entries directly (or via a holding account) where appropriate to the subpostmasters’ account and then reconciling those accounts to the branch accounts provided by the Horizon system – a somewhat basic techinque thus avoiding twenty four years plus of grief - was and remains beyond the intellect of senior finance staff in Post Office Limited.

Evidence to be given on 10th May, 2024 might provide more information.

Replies (5)

Please login or register to join the discussion.

By lionofludesch
19th Apr 2024 17:57

I have no words left.

It's impossible to imagine how this could have been c0cked up so incredibly badly on so many levels.

Thanks (3)
Routemaster image
By tom123
20th Apr 2024 18:48

Do we know much about the transactions themselves that gave rise to the "shortfalls".

Twice as many people being recorded as paying for car tax in cash as has actually happened for example?

I've never seen anything that describes what the systems thought had happened.

Thanks (1)
Replying to tom123:
By FactChecker
20th Apr 2024 20:18

No-one's saying (if anyone knows) ... but I suspect you're falling into the same trap as those benighted PO managers who expected to find a logical transactional trail and, in its absence, decided the numbers must be real (so 'no cash' = fraud).

Almost certainly (most of) the discrepancies will have been generated 'out of thin air' by processes within Horizon that simply hadn't been programmed to cope with the whole range of 'real world' scenarios that happen from time to time.

A by now reasonably well-known scenario within RTI can serve as an example:

* Jo ceased full-time employment with ER A on 12th Jan 2024 ... and commenced new full-time employment with ER B on 15th Jan 2024;
* ER A (being more on the ball than some employers) generates a P45 for Jo on the last day of employment (by processing their Jan Payroll, just for Jo, early in the month) ... but notify HMRC of Jo's end of employment in their general Jan FPS;
* Jo presents the P45 to ER B on first day of employment and ER B (also being more on the ball than some employers) manages to pay Jo for the part of Jan in their Jan PAYE processing ... and notify HMRC of Jo's start of employment in *their* Jan FPS.

So far so good (and you think so normal).
Unfortunately, for a variety of disputed reasons (possibly the order in which ER A and ER B happened to submit their FPSs, or maybe the order in which HMRC's systems randomly chose to process them), ER B's FPS ends up being processed *before* the FPS from ER A.

So what?
Well ER B's FPS contains not just 'this period' values for Jo, but also Jo's YTD values (correctly obtained from the starter P45) - and, seeing that there is a new 'employment start date', the HMRC system creates a new Employment record for Jo. However, because it has not yet been made aware of Jo ceasing employment with ER A, the YTD figures from the FPS are all 'posted' to this Employment record.
When it finally processes the FPS from ER A, the HMRC system recognises that separate Employment record for Jo - and terminates it (thereby 'freezing in' the YTD values from ER A).

[The mind-blowing bit underneath all this is that HMRC's base systems still work entirely off YTD figures - as in they basically ignore the 'this period' figures you report and work off the YTD ones, which is after all basically how PAYE works.]

So Jo now has two Employment records (ER A and ER B) which is correct ... BUT Jo's YTD values from ER A are held in both the ER A employment record and in the ER B employment record.
Result = Jo's record (as an individual taxpayer) holds both employments - and so earnings/tax/etc from ER A are actually contained twice in the individual's YTD figures!

The above is a bit of a simplification, but there are many other (fairly common) scenarios which result in HMRC's systems creating (aka inventing) entirely false numbers.

So no need for fake/false transactions - just badly programmed processes that haven't considered certain possibilities occurring and go on, using faulty logic, to say that black is white and so 'you' owe 'me' the 'missing' money!

Thanks (6)
By GHarr497688
20th Apr 2024 22:40

Am I correct that this all happened and people went to prison just because no one reconciled the transactions.

Thanks (1)
Replying to GHarr497688:
By FactChecker
20th Apr 2024 23:28

Pretty much (as far as anyone I know has been able to find out and put in the public domain).

But I'm told that in one sense it was FAR worse - the rather basic sense that it was so much closer to an electronic till than even a basic accounting system (and so was focussed on capturing the balance rather than each individual transaction).
So no audit trail of discrete transactions and neither the ability nor intention of reconciling the stock side of things.

Whether the directing brains of the outfit were capable of understanding this, they were responsible for the system being that ineffective .. having agreed (several times) to remove chunks of the original specification that Fujitsu were having trouble in delivering (again whether due to technical complexity or merely in order to stay within budgeted profit margins - who knows).

The closest analogy I've encountered was long ago - when the early parking-meters maintained a record of the last measured balance (of coinage waiting to be extracted by the collector), but with absolutely no record of the time or amount of each transaction.
On t'other hand they had no intention of worrying about the few % of 'coins' that turned out to be foreign (or even tap-washers), because there was no-one to whom the blame could be allocated ... unlike the poor sub-postmasters.

Thanks (3)