And Statistics

[andrewducker]

According to the article here, Benefit fraud and errors hit £3.1Bn last year.  Which seems like a lot of money to your average person on the street.

Of course, it then qualifies that only £1Bn of that was fraud.  Still, seems like a lot, doesn't it?  You could do quit a lot with £1Bn.  At least, I could.

But then you can see how much the government spends on benefit in the UK and see that that's less than 1% of benefit spending.

The question being - how much effort do you want to spend to get fraud down below 0.5%? 
What's an acceptable level of fraud, considering that (nearly) every method that's used to cut down on fraud makes life harder for legitimate claimants?
I wonder how much is being spent on fraud prevention, and how much it saves.

(If I've got my numbers wrong somewhere then please correct me.)

Comments

[drplokta]

The average person on the street has no idea how large £3.1Bn is. It seems like a lot, but so do £3.1M and £3.1Tn, and they don't have any real understanding of the difference between those numbers. This is a very big problem.

[drdoug.livejournal.com]

I have no idea how large £3.1bn is, and I can do sums and work powers of ten and everything. It's way larger than anything I can hold or conceptualise in my head. I have to use rote procedures and tricks to make any sense of it at all.

(Some of my favourite tricks are dividing it by the population - which gives about £50 - or by some measure of what the Government spends in a year, or the annual or total debt - the latter of which gives about 0.1%.)

£3.1m I can almost get my head around - it's within an order of magnitude of the amount I expect to earn in my working lifetime. Which is an awful lot. £3.1tn is just plain incomprehensible, and it's within an order of magnitude of the UK's GDP.

[captainlucy.livejournal.com]

I find it fairly easy to conceptualise £3.1bn (it's getting my hands on it that I'm having trouble with. I knew I shouldn't have skipped Evil classes at school, I'm sure I'd be blowing up small Pacific islands to prove a point by now) - the advantage of a childhood obsession bordering on the clinical with space and the sheer vastness of the cosmos.

The trouble I have is conceptualising just what I would do with that money. I would either think of lots of little things where an extra £20-£30k would make a huge difference and end up with about 90% of the money still to spend, or I would think of huge projects where £3bn would maybe just about cover it provided there were no unforeseen setbacks or price hikes (why hello, 2012 Olympic Games, fancy seeing you here).

[pete stevens (from livejournal.com)]

So I deal a lot with Telecity and write them some sodding big cheques, and I just about have handle the total business activity of their entire UK operation as I've seen most of it. Their total turnover is ~ £170m with maybe £100m attributable to the UK. £3bn is 30 times bigger than that. For reference, Telecity handle a good fraction of the internet in the UK, with most of the rest of it being in Telehouse (turnover £33m).

Virgin Media has a turnover of about £4bn, on about 4million customers. I'm not sure I can visualise a project the size of Virgin media though.

Cambridge University (including CUP, UCLES and the colleges) has a total turnover of £1.5bn. So I guess the combined activities of Oxford and Cambridge universities is a reference point I can just about handle.

[bart-calendar.livejournal.com]

In nearly every commercial business in the United States the loss to "shrinkage" - which is a corporate way to say "fraud" or "theft" is 1 to 5 percent depending on the type of product sold, so 1 percent benefits fraud seems reasonable to me.

[a-pawson.livejournal.com]

And to bring it into context, the HMRC's own estimate of the UK tax-gap (tax not collected which should be) is £40 billion. In other words, tax evasion/fraud is a problem 40 times greater than that of benefit fraud.

[original-aj.livejournal.com]

Back when I worked in benefit delivery ten years or so ago, not only was that the case but the return on investment for expenditure on combating tax fraud was about ten times that of benefit fraud. Trouble is, party donors don't tend to be benefit fraudsters....

[drdoug.livejournal.com]

You could do quit a lot with £1Bn.

Yes indeed! You could pay off a whopping 0.1% or so of the UK's general Government debt as at December last year!

This really should be a priority for the Coalition, along with big-ticket items like the UK Film Council (annual budget £15m, from memory - a not-to-be-sneezed at 0.0015%).

The question being - how much effort do you want to spend to get fraud down below 0.5%?
What's an acceptable level of fraud, considering that (nearly) every method that's used to cut down on fraud makes life harder for legitimate claimants?

That is an interesting question, and I suspect the answers are likely to cluster in three groups:

(a) technocrat/economist types, who'd say fraud prevention measures should be employed up to the point at which the marginal cost exceeds the marginal amount saved (regardless of the social effect of either the benefit or the fraud prevention measures)

(b) lefty types, whose basic belief is that benefits should be universal, but in a mixed society are prepared to accept some degree of targetting, who'd say that fraud isn't really that big a deal (regardless of the actual amount of fraud)

(c) righty types, whose basic belief is that benefits shouldn't really exist at all, but in a mixed society are prepared to accept some degree of safety netting, who'd say that benefit fraud needs clamping down on (regardless of the actual amount of fraud)

(d) big-p-political junkies, who have no basic beliefs, but who'd discuss the likely instant fallout for the major parties from initiatives to reduce benefit fraud

[missedith01.livejournal.com]

As a lefty type I'd like to protest at the idea that I support universal benefits. "To each according to his need", isn't it?

[tamaranth.livejournal.com]

I wonder what percentage of DWP's expenditure goes on attempting to beat benefit cheats? And what percentage goes on attempting to avoid error? and what percentage goes on producing and reading out patronising Back To Work presentations?

And given that the article mentions overpayments, I wonder how much they save in underpayments ...

[a-pawson.livejournal.com]

I'm curious, do they not attempt to recoup overpayments. An employer would had they overpaid their staff, and would most likely succeed.

[missedith01.livejournal.com]

They vigorously pursue any overpayment, including those made in error (i.e. not fraudulently), and will in most cases succeed. The law is stacked against the claimant who may wish to argue that they shouldn't repay because they did nothing wrong ... mere absence of wrongdoing is not enough to prevent collection ... neither is absence of knowledge of the overpayment - the law varies somewhat according to benefit and type of payment but even in cases of glaring official error there is always an element of whether the claimant could reasonably have realised an overpayment was being made. If the DWP think that you should have known they'll try to collect, regardless of whether you did in fact know and regardless of their mistakes.

[ami-bender.livejournal.com]

You are assuming that the main aim is to fight fraud. If i was being suspicious, I might speculate that it has more to do with discouraging people from going on the dole, or to motivate them to get off it.

Came here from a link

[rainbow-goddess.livejournal.com]

Hello, you probably have no clue who I am but I happened to see a link on Twitter to your post here.

I live in British Columbia, Canada. Around 15 years ago our provincial government decided to have a "Crackdown" on welfare (benefits) fraud. This "fraud" included someone being given a bag of groceries by their sister and not reporting the value of that bag of groceries to their case worker so that that amount could be deducted from their welfare cheque as "income." It included someone selling a chair they no longer needed and not reporting that as "income" to their case worker. One of their ways of combating so-called fraud included reducing everyone's benefits by $50 a month.

In the end, the amount the government spent on ferreting out that so-called fraud was several times more than they gained in kicking the "cheats" off welfare.

Overpayments do happen, though usually not for fraudulent reasons; it's more likely that someone just didn't understand the ever more complex rules and regulations surrounding welfare payments. The person will be expected to sign a repayment agreement, and then about $10 to $20 a month will be deducted from their benefits until the amount is repaid.

We got a new government in 2001, and that government decided to crack down on disability cheats. Their aim was to save a few million dollars by proving that not everyone collecting disability was actually disabled. The publicity around this was enormous. In the end, about 12 people lost disability payments, and several hundred MORE people started receiving them because of the publicity around what the criteria was to qualify for disability.

Re: Came here from a link

[andrewducker]

That's pretty much exactly what I'd expect. People are so worried about cheats that they make life hell for the people on benefits.

I'm curious - who/what linked to me?

Re: Came here from a link

[rainbow-goddess.livejournal.com]

The Twitter user who linked to you is called "Dr. Jon". He also linked to another of your posts about the Afghanistan WikiLeaks.

Re: Came here from a link

[andrewducker]

Cheers! The Good Doctor and myself have something of a mutual admiration society :->

Re: Came here from a link

[the-gardener.livejournal.com]

In the 1980s, I worked for the social security arm of the DHSS (as it then was). From first-hand experience, I can say that the colossal effort put into chasing down minor transgressions by people who didn't realise that they were committing "fraud", as it was technically defined, because they weren't aware of the minutiae of the rules, was phenomenal -- and quite disproportionate the the money actually saved (claimed as Weekly Benefit Savings, or WBS, and calculated by taking away the number you first thought of). It was difficult to conclude otherwise than that the whole effort was intended to put people off claiming benefit in the first place, and thus keep the claimant count artificially low in order to demonstrate that Thatcherite economic polices were working.

The coalition government will doubtless be pursuing the same fiddle. As they will have to, when the unemployment count (as both an absolute figure and as a percentage of the workforce) starts to rise during the winter, and the number of company bankruptcies mushrooms, and the economy tanks big time. (But if we have to relive the 1980s, can we at least do it without the giant shoulder-pads and the flicked hair?)

Re: Came here from a link

[the-gardener.livejournal.com]

"disproportionate the the money" should of course be "disproportionate to the money". Failure to bloddy proofread, chiz chiz

Re: Came here from a link

[captainlucy.livejournal.com]

Awww, the giant shoulder pads and outrageous hair at least gave us something to laugh at in the dark old days.

[matgb]

On the numbers...

A big chunk of benefits expenditure are non fraudable (for the most part). Pensions and child benefit are a massive chunk of that budget. While it's possible to lie about the existence of a child, or keep claiming for someone post death, I think that's rare.

So the actual fraud is from a much smaller segment of the total spend. But I don't know how big that segment is.

Regardless, I agree, I'd much rather we dealt with tax dodgers and similar. Like HMRCs landlords.

Of course, I'd rather replace the entire benefits system with something simple and universal, increase claimant rates and decrease fraud potential, but that's not the discussion.

[andrewducker]

I'm theoretically in favour of universal income (and then a flat-rate tax at about 50%), but I've never been able to make the numbers add up.