When two leading voluntary-sector bodies set about calculating the total income generated by charities in England and Wales, they produced figures so utterly contradictory that they appeared to render the studies meaningless.
Arriving at a definitive statistic was never going to be an easy task for either NCVO or the Charity Commission with accounts from 185,948 charities to sift through. There was always likely to be a gap in their findings but no one expected quite such a chasm.
Earlier this year, NCVO revealed that total charity income stood at £15.6 billion and that this had grown by 7.6 per cent since 1997. This figure was published in January in the umbrella body's latest UK Voluntary Sector Almanac.
But the Commission released its annual report earlier this month which stated the figure was in fact £26 billion and it had risen by a massive 69 per cent over the past five years.
It does not require a degree in mathematics to calculate the Commission's figure is almost twice as high as that supplied by NCVO.
Dr David Wilson, a partner at Knox Cropper, an accountants specialising in the voluntary sector, says: "Sixty-nine and seven are too far out. It appears something has gone wrong with the figures."
This is an opinion shared by many people in the sector, who feel that the publication of two such different studies is confusing. In particular the Commission's figures show a huge rise in income, which does not reflect the experience of many charities whose investment income has fallen because of recent drops in the stock market.
The Commission itself has given no explanation as to why the two figures differ so wildly. But there are at least some partial answers and they are attributed more to different methodologies than any flaws in the adding up.
The Commission arrives at its figure by a process of totting up the income from every charity. Each charity with an annual income in excess of £10,000 is required to submit its accounts, while those below this mark are asked to fill in an annual update form with basic information about the charity.
These days any anomalies involving financial figures are blamed on the accountants. However, there was never any suggestion that the Commission had manipulated its figures, so could accounting procedural changes explain the differences?
Murtaza Jessa, charity partner at chartered accountants Trustient, doesn't think so. "The introduction of SORP 2000 has made the process more transparent but would not mean higher income. There has been no change in accounting requirements that would explain this difference."
The Commission's method is simple but inclusive and misses very few organisations.
However, it fails to take into account double counting of income that appears in more than one charity's annual report. The most notable example of this is charitable trusts. Each trust's income appears firstly in its own accounts and secondly in the accounts of whichever fundraising charity it decides to support financially.
NCVO relies on sampling and projection rather than attempting to add up every set of accounts. It analyses the accounts of all charities whose annual income exceeds of £1 million as well as 7.5 per cent of charities whose annual income is above £100,000, and less than 1 per cent of those whose annual income falls below that figure.
NCVO spokesman James Georgalakis says: "We believe our figures provide an accurate snapshot of the income of the voluntary sector. The Government and the media use them when discussing the sector and no one has said to us they don't seem right."
This methodology has the advantage of looking in more depth at specific sets of accounts, which avoids anomalies such as double counting, but runs the risk of sampling errors because it doesn't study as many annual reports.
Paul Palmer, professor of charity finance at South Bank University and who helped to establish the accounting practices for NCVO Almanac, says: "You could argue that NCVO's method produces results which are nearer to what an accountant would describe as a fair picture of the sector. But both are correct, they are just calculated differently. The Commission's methodology shows the real resources that are available to the voluntary sector."
The Commission's sharply rising figure is also partly explained by the fact that there are now more charity accounts to calculate than five years ago. The figure has increased from 181,000 to 185,000, although it peaked in 2001 at 188,000.
It's also worth remembering that NCVO doesn't include churches and charitable public schools. This will have had a tangible effect as the Church of England which has large assets, tends to invest more heavily in property rather than on the stock exchange, and has been less affected by the recent fall in the FTSE Index.
But while there are mitigations for having two such contradictory figures, they hardly justify the confusion. It leaves everyone scratching their heads trying to fathom out who to believe.
Charities Aid Foundation's communications director Neil Jones says: "When people ask me what the income of the sector is I tell them between £15 billion and £20 billion.
"The Commission's figure does seem rather high. I haven't met many people who say their income has grown by 69 per cent over the past five years."