Free Markets, Free People

newspaper circulation

Newspaper print circulation is down about half in last ten years

I did a couple of posts back in 2012 and 2013 about newspaper print circulation at major newspapers, compared to 2004. Seeing last year’s circulation figures made me curious about how things stand at the ten year mark. Here are the results:

 

Newspaper 2004 2012 2014 +/- %, 10 years +/- %, 2 years
USA Today 2192098 1627526 1156871 -47.23% -28.92%
WSJ 2101017 1499204 1256292 -40.21% -16.20%
NYT 1119027 717513 680905 -39.15% -5.10%
LA Times 983727 489792 405213 -58.81% -17.27%
Wash Post 760034 434693 399757 -47.40% -8.04%
NY Daily News 712671 389270 313178 -56.06% -19.55%
NY Post 642844 344755 261312 -59.35% -24.20%
Chicago Tribune 603315 388848 324620 -46.19% -16.52%
Denver Post 340169 236223 188630 -44.55% -20.15%
Newsday 553117 278369 247703 -55.22% -11.02%
Houston Chronicle 549300 234483 225032 -59.03% -4.03%
Dallas Morning News 528379 345342 172690 -67.32% -49.99%
SF Chronicle 499008 229176 145520 -70.84% -36.50%
Arizona Republic 466926 274783 244726 -47.59% -10.94%
Boston Globe 446241 180919 159458 -64.27% -11.86%
Tampa Bay Times 348502 299393 217597 -37.56% -27.32%

As I explained in the previous posts, I focus on print circulation because, for major newspapers, that’s where most of the money comes from. Newspapers do get money from the web, of course. However, most of them have minimal web-only subscription revenue, and their advertising dollars on the web are only about 15% of their print advertising revenues and growing slowly according to Pew Research. That same report shows that overall advertising revenue (including online advertising) is down just a bit over 50% for the 2004-2013 period.

I ignore the web “circulation” numbers touted by newspapers, because they’re meaningless without a complete explanation of how they were measured. Unique visitors for the year? Well, people have multiple computers, and they clear their browser cache sometimes. Even when an explanation is given, those numbers can be gamed in various ways. The money is what counts, and newspapers have struggled to increase the amount of money they get from web publication over the last six or eight years. There’s no indication they’ll solve that problem.

Doing a bit of math on the above numbers, the drop in the aggregate circulation of these newspapers combined from 2004 to 2014 is just over 50%. Aggregate drop from 2012 to 2014 is about 20%.

Many dissipative phenomena in the real world have an approximate exponential decay shape to the graph. That is, the newspapers might lose, say, 10% of their readers each year, but that 10% is a lower number each year, so the decrease flattens out in actual counted numbers. That’s my best guess for the near term future of circulation for major newspapers.

However, dropping revenue also affects quality. This hit my hometown newspaper, the Tennessean, at least ten years ago. You could see it exposed unambiguously in grammatical and printing errors. I also think the quality of the articles dropped to the point that I wasn’t willing to invest time in reading them, but that’s a more subjective judgment. Except for local events such a major water outage last year, I don’t pay any attention to the Tennessean.

When that happens, the days of a newspaper are numbered. They enter a vicious cycle in which more people drop them because of their marginal or poor quality, and that erodes revenues further, which erodes quality further, and so on.

There’s no obvious way to reverse any of that, no matter how innovative they get on the web. Advertising revenue for want ads isn’t coming back; Craigslist and its smaller relatives have captured it and I see no way for newspapers to get it back. Not even middle aged people get newspapers for movie ads anymore because they can find anything they want to know on their phones immediately. Retail advertising continues to suffer as retail closures start to impact suburbia, and dead malls continue to pile up.

So, with that dead horse beaten to a pulp, what are the likely effects outside the newsrooms? 

Right now, the New York Times and the Washington Post continue to have an outsize influence on political thinking. I don’t think either one is going to vanish any time soon. The left will no doubt find the Times so indispensible that it will find the money somewhere to keep the lefty editorial outrage and the slanted reporting pouring out of Times Square and setting the agenda for TV news reporting. The Post, under Bezos, seems to be becoming marginally more balanced, which is a good thing.

The Wall Street Journal maintains a decent hold on center-right readers, though it’s a lot more center than right these days. As the only major newspaper I read with any frequency (couple of times a month) I see the quality dropping. But for now it seems financially stable.

Almost all the others, though, are in trouble. I have to wonder if the recent successes of the GOP at the state and local levels have not been facilitated to some extent by the lack of effective opposition from the typically-liberal local newspapers. The fewer people who read them, the less able they are to torpedo Republicans and shield Democrats.

Naturally, you don’t see a lot of reporting on all this in the media. They don’t have much interest in exposing their own weakness. The reporting they do typically touts “total circulation”, which means they get to include their gamed web numbers. USA Today also started an insert program with a lot of local newspapers, so they like to pretend that this is equivalent to regular circulation. It’s rare for any of them to make their print declines front and center.

The main lesson here is that limited government types can afford to stand up to these biased media types more each year. I think that’s more true at the local level right now, but I also think there are a lot of people out there hungry to see the left-liberal twits of the major national newspapers put in their place as well.

*** Update 5 April 2015 ***

It occurs to me that, if the decay in readership of major newspapers is really a bit similar to exponential decay processes such as radioactive decay, then ten years would be the half life of newspaper readership. We might then use that half-life as a rough-and-ready estimator for future declines. It would suggest that by 2024, the newspapers will have lost around 50% of the remaining readers, and be at 25% of their 2004 readership.

Naturally, there are too many real-world factors to put much confidence in such an estimate, mainly because of the “death spiral” end game for such businesses. But it’s still an interesting first cut way to think about it, and it might help us detect the death spiral start point.

Top ten newspapers and their circulation–a six month update

Six months ago, I did some numbers and commentary on declining newspaper circulation.* By chance, I noticed a couple of web articles that give some more current numbers, so I decided to revisit the older article and see how things are proceeding.**

Here is a table I created with print numbers from 2004, mid 2012, and late 2012-early 2013 for the current (2013) top ten newspapers. The current top ten list is taken from an AP article on Huffington Post, and is ranked by their current circulation.

 

Newspaper

2004

2012

Early 2013

6 mo +/-%

Total +/-% since 2004

WSJ

2101017

1499204

1480725

-1.23

-29.52

USA Today

2192098

1627526

1424406

-12.48

-35.02

NYT

1119027

717513

731395

1.9

-34.64

LA Times

983727

489792

476148

-2.79

-51.60

Wash Post

760034

434693

432454

-0.52

-43.10

Chicago Sun-Times

453757

361523

392889

8.68

-13.41

Chicago Tribune

603315

388848

368145

-5.32

-38.98

NY Daily News

712671

389270

360459

-7.40

-49.42

NY Post

642844

344755

299950

-13.00

-53.34

Denver Post

340169

236223

223871

-5.23

-34.19

 

Back in November, I said

USA Today looks vulnerable to me, because it looks like the easiest national newspaper to replace with a web-based aggregation app. They do very little original reporting except for the sports section. They have not yet ramped up a decent web presence, and it’s pretty late in that game.

In the latest numbers, USA Today’s print circulation is down a staggering 12% in just six months. I’d like to say I was prescient, but that’s so much, I suspect that the data isn’t comparable. I suppose it could be correct, especially if they lost a major hotel chain or two as a distribution channel. It does seem indisputable that they they are on a long term trend of losing circulation fairly rapidly.

It appears that USA Today did ramp up their web presence somewhat. The reported number of “web subscribers” went from about 86,000 to 250,000. I suspect they’ve started counting the numbers differently; that much increase out of the blue, with no special reason for more eyes on their site, looks unlikely. Since they have no paid web subscribers, it almost doesn’t matter anyway because the revenue from web advertising isn’t going to support their current business model. (The uncertainty about web numbers is one of the reasons I think the methodology might have changed enough to make the print comparisons suspect.)

I also noted circulation alarms for the Washington Post last time:

The Washington Post looks vulnerable too. It also has limited web presence, and print circulation is down a staggering 40%+ in eight years.

The six month circulation change isn’t too bad for them, but Ed Driscoll noted yesterday that their financials have taken a big hit in that time period. Their earnings are down 85%. 

There are a couple of reasons I don’t pay much attention to the web numbers. First, it’s hard to compare the numbers or get any idea of trends without details on their methodology for counting “subscribers”. For example, they could take the count of people who have gone through a silly registration process where they ask for an email. Someone might register that way for one article and never come back. Or it might be based on visits, but there are lots of ways to fudge those, depending on how you count and define things.

Second, I’m guessing they are using a methodology that’s favorable to their numbers, and they still lag. For example, the largest reported number of web subscribers by any of the majors is about half of what the Drudge Report gets in unique daily visits. Drudge’s monthly unique visits would make that ratio fifteen to one instead of two to one. I mentioned last time that Huffington Post has passed NYT in daily visits.

The overall story means steadily decreasing revenues for everyone except possibly the Wall Street Journal and the New York Times, and I wouldn’t be too sure about the Times. Occasionally a regional paper will have a good run and make up some ground, as the Chicago Sun-Times has done recently (passing the Tribune on circulation in fact), and I called out the San Jose Mercury News for impressive growth in the last post. But those cases are rare, and don’t seem to be long lasting effects.

Way less money to spend is convenient in some respects, though. It’s easy for an editor to rationalize ignoring a complex story such as Benghazi. Unconsciously, he may not want to cover it because of the danger to his precious historic president, but he can tell himself he just doesn’t have the resources.

One of the messages the right needs to communicate and make part of the popular understanding is how declining revenues have constrained the reporting at major newspapers. That would be one way to explain to people, without getting partisan about it, that those newspapers shouldn’t be regarded with the authority most everyone gave them thirty years ago.

 

* The Washington Post link in that blog post, showing 2012 circulation figures, is dead now. It was apparently based on an AP story, and got removed after a while. I found the original AP story on Yahoo, with all the numbers from the original cite. It’s here.

** I should repeat the same caveat as last time: I am looking primarily at print circulation declines, and so I have to do some arithmetic because the newer numbers combine web and print. Those numbers also give the web number, so I subtract to get the presumed print circulation. It’s possible that I’m misunderstanding what the web numbers mean. Some of the “web subscribers” might also receive a print edition. In that case, the print numbers would be higher. But since I think the industry would want those numbers to look as high as possible, I don’t think they’re defining things that way.