Everyone in the newspaper industry has been pouring over the latest circulation figures, released last week. They seem to offer some comparatively good news for Fairfax broadsheets. The weekday Age rose very slightly Monday to Friday, fell a little on Saturday and rose a little on Sunday. The Sydney Morning Herald rose slightly across the week.
Since Audit Bureau of Circulation rules changed a while ago, most of the industry has accepted that the quarterly figures are a fair dinkum way of measuring what advertisers are getting for their dollar, and the success of journos and editors in attracting readers. But are they?
There are two issues regarding Fairfax’s figures in particular. First, I am reliably informed that it is common practice within Fairfax to slightly “under declare” the real figures during periods of growth, so as to avoid “bubbles” that can make future figures look bad. In other words, there are usually a few thousand in circulation figures “in the bank” that can be used to disguise any uncomfortable fluctuations. Which means we don’t really know whether circulation is going up or down in any particular quarter.
The other thing we need to understand is the impact of bumper editions — those occasions, usually on holiday weekends, in which basically the same paper is distributed on both Friday and Saturday — same sections, same ads, only the news stories changing.
Sources inside News Limited are rather contemptuous of Fairfax’s addiction to bumper editions. But doubtless Rupert has his own ruses.
The Audit Bureau allows only four bumper editions a year, but they nevertheless distort the figures.
It works like this. On a normal Saturday, The Age would sell around 140,000 copies to home delivery subscribers, plus another 150,000 odd to casual buyers. That makes total circulation of about 290,000.
But if there is a special bumper weekend edition, the same paper also sells an additional 140,000 “repeat” copies to home delivery subscribers, who get the paper on both Friday and Saturday. That 140,000 repeat sales gets counted as additional Saturday circulation even though the paper comes out on Friday.
Average it out over the 13 weeks of an audit period, and you boost the circulation of Saturday’s paper by more than 10,000 — yet because the advertisements don’t change, the advertisers get no more bang for their buck.
Now, in the figures released last week, there was one bumper edition included — the one over Christmas.
In the next figures to be released, covering January to March, there will be two — New Years Day and Australia Day. Thus the Saturday circulation in that quarter will be boosted by more than 20,000, depending on how much of that circulation is kept “in the hat” and undeclared to flatten out any bad figures ahead.
The problem is that once indulged in the bumper edition padding, it is like a hard to quit drug. Last year, the figures for the first quarter of the calendar year also included an Easter bumper, because Easter fell early in 2008.
This year, Easter falls into the next quarter, so it will boost the April-June figures — expect to see a rise in Saturday circulation compared to the same quarter last year — but leave a hole in the January-March ones.
The decision on whether or not to have bumper editions is driven, I am told, far more by whether there is a need to boost circulation figures than by any consideration for the needs of readers, or, for that matter, newsagents.
What it really means is that the circulation figures are at best a heavily spun and manipulated version of the truth. While really big rises and falls over longer periods cannot be disguised, there really is no point at all in crowing or mourning over variations of less than, say, 2%.
And that’s even before we get into the impact of ultra-cheap copies and give-aways.
So the only thing that can be said about last week’s release of figures is that for Fairfax broadsheets, there is no information that can be relied upon.
On the other hand, the big declines of circulation in News Limited tabloid newspapers across the nation are probably at least a version of the truth.