Cast your mind back to the back end of 2012 when the big retailers and banks were staging auto-enrollment.

The big employers were spending hundreds of thousands of pounds re-coding their payroll software or were ransoming themselves to “middle ware” – which would soon be renamed “muddle-ware”.

There was a need for a data standard to which payroll software suppliers could build. The standard could save payroll millions and this could be passed on in lower pension integration costs to employers. A group of software suppliers , bureau operators and some pension providers started meeting , convened by Andy Agethangelou’s Friends of Auto-Enrollment and led by Will Lovegrove of PensionSync , watched over by Neil Esslemont of the Pensions Regulator.

They devised a data standard, it was called PAPDIS, but it arrived too late. The PAPDIS data standard was rejected by Nest on the grounds that Nest considered itself the data standard.

Fast forward six years and in the meantime those big red spikes of companies staging from 2016 to 2018 found a way. Payroll software companies found a way and the pension providers found ways to integrate. The great vaccination  – the API is finally happening. But it is happening selectively

PensionSync integrations

AEclipse integrations

If you are a small employer, choice is still governed by the interoperability of payroll and provider and the mainstream providers are those who have invested in interoperability.

Why did they do so? Here is Dave Lunt , commenting on linked in on behalf of People’s Pension.

Less cost for People’s Pension means prices stay low. Better data quality means less problems at “claim”. More security means less scamming. But how much has been lost by our failing to get data direct from payroll submission. When she worked with PensionSync , Ros Altmann made her feelings clear

As chair of @Pensionsync I have been amazed at the scale of data errors coming through from payroll to pension providers. Why aren’t there requirements for accurate info to ensure employers pay the right amount? If they pay too little, who will know?! No proper checks

— Ros Altmann (@rosaltmann) April 4, 2019

In truth , auto-enrollment muddled through and the great Government success story has been achieved with some grievous failures from employers who have over -or more seriously – underpaid contributions. Many payments got lost (we will not forget the cost of rectifying NOW’s middle-ware). many employers chose sub-optimal workplace pensions to suit the needs of their payroll.

Let’s not suppose that auto-enrollment gave us all the answers. But it did at least teach us where we shouldn’t be going wrong.

Lessons for the future

If you look at those advertisements of AEclypse and PensionSync you can see that what matters to future purchasers is integration to some core providers and both products are being targeted at a relatively few payrolls.

Note to the Pensions Dashboard Programme (1): a market will form around the major participants. The Data Application (go live) point is when the market feels that the service is sufficiently inclusive and auto-enrollment shows there is no appetite for 100% inclusion.

Note to the Pensions Dashboard Programme (2); direct integration can involve competition between integrators as displayed but at great cost. It would have been better off for all if a data standard had been adopted when auto-enrollment was young. We do not want to see the same mistake made twice. The pensions dashboard needs the dynamism of the private sector and the co-operation of Governmental organisations to common purpose. That’s why I’m a supporter of the approach being taken by the Pensions Dashboard Committee in getting the common data standard agreed at outset.

Note to the Pensions Dashboard Programme (3); Pensions are payable because people defer present pay for future pay. Pension providers do no more than maximize the effeciency of that deferral. Auto-enrollment is a slight of hand that enables this deferral to happen without compulsion. But auto-enrollment works because savers stay out of the way, it does not encourage savers to get involved with their pensions. The pensions dashboard seeks to reverse that process. Having effectively excluded the saver from the process, we should not expect the saver to be well informed. We need some simple ways to get people engaged with their money. Auto-enrollment data integration has worked where the service has been direct to the customer.

Pragmatic on inclusion, authoritative on data standards and insistent on consumer access to their data.

Auto-enrollment roughed out its solution and it’s not a perfect solution. But it’s got there.

The pensions dashboard was conceived as combined pension forecasts in the early years of the millennium.  AE is about real money, the dashboard is about the data that tells us the money is real.

The dashboard has been slower because money makes the world go round. But … the money we are talking about is our money!

Why can’t we integrate to our pensions?

Those who have managed our money, have held our data and held it close. Rob Mann is not the only member of the public who is frustrated by why his data and his money aren’t more easily available.

If I could get a transfer value by email in less than a week that would be a start…

— Rob Mann (@xx213) August 8, 2020

It is a crying shame that the priority for pension providers has been  in integrating  with payroll – not with dashboards.  But that is because money talks. No.w we must make our money talk. We must demand better access to our data and our money. We cannot allow the dashboard delivery date continue to recede into the distance. We cannot allow our providers to dictate the service we receive and leave these questions unanswered.

Google “pension integration” and you will see hundreds of images of flow-charts. You have to scroll a long way till you get to a human face that isn’t being used to sell software. When you do , this is what you find.

The other face of “pension integration”