Pretty soon, proponents of free digital news will have to own up to the implications of their model.

The structure is flawed. To rely on online ads as the sole source of revenue is both unsound in theory, and in practice it’s having disastrous consequences in regional newsrooms, mostly notable at the Times-Picayune and other Advance Publications papers across the South.

When I wrote my post a couple weeks ago—in the wake of the Journal Register Company’s second bankruptcy—I wondered if I had missed something. Was there a hole in my argument?

I said that the free online game is a volume game: the more posts a news organization publishes, the more hits it will generate, the more chances for something to go viral, the higher ad rates it can charge. To re-quote Felix Salmon:

When you’re working online, more is more. If you have the cojones to throw up everything, more or less regardless of quality, you’ll be rewarded for it — even the bad posts get some traffic, and it’s impossible ex ante to know which posts are going to end up getting massive pageviews. The less you worry about quality control at the low end, the more opportunities you get to print stories which will be shared or searched for or just hit some kind of nerve.

Put another way: The free structure carries with it an incentive toward volume that is built into the model. The bias toward quantity is baked in.

And quantity, given limited newsroom resources, is the enemy of quality.

The thing about structure and incentives is that they don’t make certain outcomes inevitable. They just make them much more likely.

You can listen to me, or you can just read this incredible SF Weekly piece on SEO-happy Bleacher Report:

Sample:

Perhaps uniquely among journalistic entities, Bleacher Report has a “blanket policy” forbidding its writers from seeking out and breaking news. A dictum on the site states: “While we don’t doubt that some B/R writers have contacts they know and trust, a problem arises when we’re asked to take a leap of faith that those sources are both legitimate and accurate.”

Under the free model, that makes perfect sense. (Ryan Chittum further explores the implications.)

On the other hand, I argued, a paywall carries with it different incentives. The new revenue stream can be used to help maintain quality, but at the same time, the paper is incentivized to produce journalism that someone might actually want to pay for. In our new disaggregated news environment, when newspapers can no longer rely on omnibus models that had people buying the paper for many reasons (news, horoscopes, grocery coupons, etc.), the paywall can be seen as even more a referendum on quality than in the print days, though I admit the decision to buy a newspaper is complicated. Of course, paywalls don’t guarantee quality. Even Gannett has them.

I appreciate that Mathew Ingram took on the idea. I want to know how “free” actually incentivizes quality. But, really, instead of argument he offers a misrepresentation of what I said, and a rather crude one at that:

But do paywalls automatically mean that you get better journalism? In other words, does a free and ad-supported model mean that the journalism you get is of lower quality, because of the “hamster wheel” effect?

Lyn Headley caught the device right away:

Starkman is trying to look at incentives, and the way they encourage behavior. He’s not arguing that these incentives (page-view counting and things like it) directly result in the behavior (short, limited pieces and frequent publication, aka “hamsterization”), only that they are compatible with it and encourage it. An incentive is like an angel or a devil on the producer’s shoulder, whispering in her ear. Ingram senses, I think, that Starkman wants to show a stronger link than mere association between the incentive and the behavior, and also that Starkman has not shown the link. So he characterizes Starkman as asserting that the incentive “directly translates” into the behavior, whereas Starkman uses other summary phrases (e.g. the behavior is “a logical, if not inevitable result” of the incentive).

Ingram posits a similar straw man, that someone supposedly is saying, “That simply because your content happens to be supported by advertising, that it can’t be of high quality.”

But of course no one says that.

The question, as I said, twice, isn’t whether a structure requires something, but does it encourage the thing? And while we’re dueling, my argument might be bogus, but it is not, strictly speaking, a logical fallacy, “a failure in reasoning that renders an argument invalid.” The reasoning is all too valid; that’s the problem. Free online is driven by volume.

And citing, as Ingram does, Huffington Post and The Washington Post absolutely does not invalidate the reasoning.

First, WaPo the newspaper, with its free online model, is a huge money loser, subsidized by other parts of the business. So whatever quality it is still able to produce in no way can be attributed to the free model. It comes in spite of it.

Rather, the paper and its declining journalistic and financial fortunes is a cautionary tale, a living example of the dangers of clinging to the mast of free. For more, read Ryan’s “The Washington Post Co.’s Self-Destructive Course,” a deconstruction of the Post’s free news strategy.

Put a third way, if Clay Shirky—Mr. Here Comes Everybody himself—says the Post, “should turn to their most loyal readers for income, via a digital subscription service of the sort the Times has implemented,” that should at least give free-news advocates pause. The Shirky case is nuanced, but there it is.

And, because HuffPo laudably does some great longform and investigations, that doesn’t mean that volume-based incentives of its free model don’t exist. It means it’s doing this work in spite of its model. As I said in my post, “Low High-volume low-quality isn’t inevitable under the free structure, but the model’s incentives run in that direction.”

So, what is the quality argument for choosing it, particularly at a regional newspaper today?

(And Ken Doctor makes a detailed case that even the financial argument—forget the journalism argument—is dubious at best over the long term for Advance in New Orleans.)

Mathew now points to Talking Points Memo’s membership system as preferable to a paywall. I’d say that asking for money to pay for additional content seems like a paywall by another name. But even if it’s different, this is another vote against free news and for charging readers. Indeed, Josh Marshall says he needs to charge so the site can do in-depth work.

This is one of the big reasons we’re doing Prime. We’d love to publish those [long] pieces. And we know there are a lot of our core readers who’d love to read them. But the economics of our company is based on reports that get lots of readers. If we spend a month or two producing a 10k or 15k word mini-book and we put it up on TPM and a few thousand people read it, that’s a big problem for us. So Prime is about setting up a business model within our larger business model that allows us to do something like TPM Singles. We don’t want to not do those pieces just because they probably can’t generate the mass audience that will pay for them with display advertising.

Again, under the pay model, that makes sense.

I think Mathew has a fair point when he says that at newspapers, ads have traditionally carried the revenue load. It’s true, as he says, that “[f] or the most part, advertising has paid the freight for journalism for decades, just as it does online…” and that “…newspapers have always been driven in part by a desire for advertising revenue, even if they charged a small fee to readers.”

But that’s just it: it’s a new day. I’m suggesting the mix needs to change.

Relatedly, Jim Brady offers a spirited defense of the journalism produced by the all-free Journal-Register Company against a pointed takedown of by the Awl.

The details are beyond the scope of my argument today, but Brady is on the right track by at least pointing to actual examples of the Journal Registers’ work, allowing readers a chance to judge for themselves the merits of stories cited as the best. And he and others candidly concede, the websites of JRC are “subpar” but say they will get better.

Fair enough. (The only caveat I’d mention is that I’m not sure JRC management fully appreciates the disconnect between its boastful and aggressive rhetoric about its digital orientation—I’m not naming names, John Paton—and the actual digital product a company called Digital First produces. This isn’t to say, rush the fixes. It is to say, that Awl piece didn’t come out of nowhere.)

But this debate is bigger than that. As (name-drop alert) Walter Lippmann said in 1920: “It’s not enough for [journalists] to struggle against great odds, as many of them are now doing, wearing out their souls to do a particular assignment well. The philosophy of the work itself needs to be discussed. The news about the news needs to be told.”

And the news about the free news model is not good; its flawed logic is catching up with it. Every day it seems some different organization is reversing the panicky decisions made a decade ago and trying to figure out a way to get readers to contribute more for their news.

Now, it’s the Post and Courier of Charleston, S.C. As Doctor, an honest broker in this debate if there ever was one, says:

Circulation has turned from a means (getting ad-rich papers to shoppers) to an end unto itself, actually getting readers to pay a significant share of the journalism costs. It’s a simple proposition: You ask the people who really value you and your journalism to pay you more. Surprisingly to some, it looks like many of us are willing to. Why didn’t we think of this earlier, before the carnage of cuts overwhelmed the profession? Call it a brew of misunderstanding the digital transition, of timidity, of Steve Jobs’ iRevolutions…and of desperation.

Why didn’t we think of this earlier? FON-y you should ask.

I’m sure there actually is a serious case to be made that free and quality can run in harmony, rather than at odds. But that case would have to take into account metrics other than click volume—perhaps it’s increased time spent on the site, or goodwill, or something else that would provide publishers with some pricing power over ad rates. Maybe it helps with conferencing revenues, for all I know. Ingram doesn’t make that case. If someone else has, I’d like to see it. And what’s more, I’d be happy to see it.

It’s not about the model itself, but what the model produces and what it is designed to produce.

I say the free news model is designed to produce volume. Who wants to seriously argue otherwise?

If all this were an academic parlor game, that would be one thing. But there’s a journalistic train wreck going on right now in New Orleans and in Alabama that directly attributable to the free digital model. Just last week, a reporter I know has, like other great reporters, turned down an offer to stay at the Times-Picayune under its new, free-centric configuration. The reporters know; they’re voting with their feet.

And did I mention Advance owns the Plain Dealer, the Star-Ledger, and The Oregonian?

It’s easy for us up here to advocate one model or another.

It’s the reporters, and the public, that have to live with it. Until there’s a quality argument for the free model, newspapers should avoid it.