Continuous data matching will cut benefit fraud

In this computer age, why do we still rely on claimants to tell paying agencies what money they are receiving from other parts of government when the data is already on government computers?
 
I asked that here, and North Northwester has provided this thought provoking reply:

Some shared computer facilities do exist - the local authority in my area can access a database that shows Income Support, Employment Support Allowance, Incapacity Benefit, Jobseeker's Allowance, Disability Living Allowance, some Bereavement benefits,and Working Tax Credits and Child Tax credits, though not the ubiquitous Child Benefit. It makes day-to-day assessment very much easier to do, and fraud and error easier to check...Once decision-makers have a legitimate reason to check.

My deep-cover vole at The Housing says that Local Government administered benefits assessors do have ready access to all this information - if they have a reason to go into the records, such as checking partial information about a change of circumstances received from tenants, council tax payeres, or from other agencies.

What they are NOT allowed to do is to check benefits and tax levels on a regular basis unrelated to incoming information.

Assiduous decision-makers therefore can't set themselves diary dates to check the incomes of serial failure-to-reporter who incur overpayments once or twice or three times per year.

Such a routine would save taxpayers and benefits recipients alike from fraud and error [many overpaid benefits claimants are just hopelessly disorganised or stupid rather than active and knowing embezzlers]. But it is deemed to threaten their privacy by the data protection gurus. The same local authorities that are allowed to put electronic chips in your dustbins to see what you're throwing away preparatory to fines and Green-lectures can't check repeat-offenders' tax credits. This is the most common change of circumstance that is reported to town halls, and is triggered by increased earnings or new children or partners moving into the claimant's home.

So the answer is we rely on them reporting it becasue the State doesn't want the benefits decision-makers to play Big Brother.

Doh.

If this is right, an "independent" (read "unaccountable") official is costing you and me hundreds of millions of pounds a year (benefit fraud runs at £3.5bn annually). It should in principle be easy to write rule based computer programs to throw up warnings when one of a claimant's benefits changes.

Note that this wouldn't require human eyes to be continually inspecting the claimant's records. No one would need to look, until a change triggered a warning. Decision makers wouldn't be playing Big Brother - but our expenditure would be better protected.

This is one step on from more frequent random human reviews - which can only ever examine the benefits of a small fraction of claimants. But it would require initiative from government, so I'm not holding my breath.

A more radical solution would be to follow the line of the recent Duncan Smith report and cut the number of benefits to two, from the current total of around 50. Yes, 50. But until we get there, we need effective policing of the money being paid out.

If you want means tested state benefits, you should have to accept that officials can monitor your records confidentially as they choose. If you don't want benefits officials to be able to look at your benefits data, don't claim the benefits.

Of course that will only make a dent in the problem - though a significant dent. But government should make a strategic choice: slash the number of benefits, or make benefits records subject to continuous rules-based monitoring?

But we recognise that government is about ducking unwelcome choices.