The Rationality of One Star Reviews | Hacker News Reader

The Rationality of One Star Reviews

danariely.com

18 pointssiglesias35 comments

Dan Ariely makes a baffling argument that ebooks should somehow cost even more than physical books, and that it's "puzzling" how much attention people pay to price.

It's hardly surprising that consumers would respond to what they see as unfair pricing models in whatever method is available to them.

When the easiest and most available method of feedback is through a star rating system, naturally people will use that for anything they feel the need to say, even if that's about the price rather than the content of the book.

Rather than moaning about people's desire to see a discount when they're getting a product that's cheaper to produce, Ariely should consider solutions to the issue he raises.

If you don't like people talking about the "wrong" thing in a review, perhaps the answer is to make it easier for them to talk about those things somewhere else. How about a "why I'm not buying this" box? As a onetime publisher myself, I would have loved to know WHY people didn't buy, not just the fact that they didn't.

His argument is that e-books have greater utility than hardcover books. In general, consumers are willing to pay more for products that are more useful to them. How is that so hard to understand?

He's an economist. This is a thought exercise that illustrates an apparent paradox in pricing. It's actually an extremely interesting insight for entrepreneurs (and others) who would like to set the most effective price for their product.

What's the insight? It may make sense from the economic rational-actor perspective, but the real world is a lot more complex than that.

We're not (all) rational robots performing relatively complex math to determine our tolerable enjoyment-utility per hour thresholds.

There's no paradox in expecting cost savings to be reflected in retail pricing.

The idea that an ebook has more utility than a real book is hardly uncontroversial. You can't read it without the appropriate technology and a power source, it can't be easily lent to another person, can't be resold, and lacks all of the physical qualities that make a well-designed book readable, such as paper quality and typography.

The advantages of ebooks are that they are easier to carry, can be purchased instantly without a trip to the store, and are never out of stock.

It's the convenience of ordering that has made them popular, but customers aren't idiots. They know that printing, binding, shipping and storing books is expensive, and that those costs are reflected in the price. When you remove all of those elements, they expect the price to go down as well.

How is that so hard to understand?

But you just re-stated the entire article -- that, despite the obvious increase in utility afforded by e-books, customers are not blind to producer prices and expect (nay, demand) fairness in pricing.

Yes, this is common sense. But it is "paradoxical" (or lets' just say, confounding) from the perspective of traditional economics -- i.e., it is hard to construct a model which accounts for this behavior if you consider consumers as rational actors. The behavioral economics movement -- of which the author of this article is a member -- came about to highlight the ways that human behavior is not "rational" in the traditional economics sense, and concepts such as perceived fairness or context have a huge impact.

As an aside, I didn't really read his last paragraph until just now, and thus I didn't really understand the undercurrent of anger in the comments here. I do agree that it's a foolish conclusion he makes (I'd summarize it as "bend over") and the whole point I walked away with from his previous discussion was that producers cannot expect to set prices based solely on utility; they must take into consideration perceptions of fairness and producer costs.

I believe I'm in violent agreement with your second paragraph. A mismatch between the predictions of an economic theory and what happens in the real world demonstrates a predictive (and therefore descriptive) inadequacy of that theory, rather than a (moral) failure of the world.

Sorry, but you insist on phrases like "obvious increase in utility" when that is not at all the case, as I just pointed out.

I could just as easily talk about the "obvious increase in utility afforded by real physical books."

This isn't even a case of rational or irrational behavior. It's a case of an economist misunderstanding the real value of a product to consumers.

Just because people buy ebooks doesn't mean they view them as having greater utility than real books. Does a movie from iTunes have greater utility than a DVD? An MP3 versus a vinyl record with beautiful sleeve art, gatefolds, and liner notes? Is an iPad "magazine" more valuable than a physical one?

Digital simulacrum of real-world objects are popular because they can be purchased on a whim, and produce instant gratification. That doesn't make them inherently superior.

One author friend of mine, Margaret Weis, has seen ebook sales outpace her physical book sales this year. At the same time, her physical book sales have not suffered at all. What has happened is that people have been buying ebooks for instant gratification, then going back and purchasing the real thing later. That would seem to speak directly against your opinion that ebooks have an "obvious increase in utility."

You need to reexamine the Amazon customer comments that this article is based on. Nobody is saying that ebooks are less useful therefore they should cost less. If that were the case people would simply purchase the hardback copy and be done with it. This is simply outrage over the "fairness" of charging more for something that is cheaper to produce.

Here's my evidence:

Sunk cost fallacy: http://cl.ly/1V0Y2L3W2Y0l3A3S3F28 Fairness in cost fallacy: http://cl.ly/0i3u0V183e0g273l421e, http://cl.ly/3s1Z3l3w06291g0N0W15

Who is saying that the Kindle version is worse?

But it is "paradoxical" (or lets' just say, confounding) from the perspective of traditional economics -- i.e., it is hard to construct a model which accounts for this behavior if you consider consumers as rational actors.

Actually it's easy, and the last comment on the article nailed it. Ebook pricing is an iterated "ultimatum game": http://en.wikipedia.org/wiki/Ultimatum_game. A copy costs the publisher essentially nothing, creating a large amount of economic surplus to be divided between the seller and buyer. By setting the price higher than a physical book, the seller is attempting to capture almost all of the surplus, similar to keeping $99 and offering the other player $1 in the ultimatum game. In both cases if it's the only offer you're ever going to get, it's rational to take it because it's the best you can do. But in reality there will be future interactions with sellers. Refusing a slightly beneficial deal (and encouraging others to do so, in the case of reviews) will likely encourage them to offer more favorable deals in the future, producing a net benefit for buyers.

Thanks for that. It's good to see some rigour being brought to something that's hopefully intuitive (long-term gain).

The idea that information, such as ebooks, should be priced based on consumer utility rather then marginal cost is something that is a naturally attractive viewpoint for the seller but not necessarily for the buyer. When you have a relatively high initial sunk cost and a relatively low marginal production/distribution cost, a lot of sellers will be trying to work very hard to push value based pricing since the alternative, of a small profit over marginal cost, scares a lot of them since the marginal cost is 0.

I think that consumer acceptance of this is based on a sense of fairness combined with knowledge of the sellers cost. A lot people have no problem paying 10c for a text message, because the amount is tiny and few know that the marginal cost for the carrier is practically 0. The lack of knowledge as well as the low price (even lower for bundled plans) makes it just easier to pay rather then protest. For ebooks a significant amount of people might have an intuitive (though possibly flawed) understanding of the marginal cost and have a reason to spend time protesting, because the amounts are significantly higher then 10c.

As with any such technology driven changes in the past, I think in the long run the price will still trend more towards the marginal cost, ie 0, in the long run. Given enough time, I think that publishers and authors will realize that given the ease of distribution for ebooks they can maintain or even increase their profits by lowering unit costs and making it up on volume. This will happen whether the producers want it or not, as they will be forced to lower prices by rampant piracy.

P.S. I have to give credit for a lot of the above ideas to a great book, Information Rules, that has been around for 10 years or so.

There is utility, and then there is liquidity. Believe it or not, digital copies are less liquid than hardcopies because they cannot be resold.

I own many specialist technical books in the $70-90 price range as well as art books in the $100-150 range which I would never even thought about purchasing if I knew they could not be resold. I suppose many consumers think this way as well.

Liquidity is a very good point. I'd also include another issue which I don't have a word for, because economics is not my field.

That's the issue of permanence and authenticity, the value of knowing that the object you are looking at is in fact the same as it was when originally created, and not a reproduction or altered version.

I collect first editions and other vintage copies of magazines I admire. Among my collection is a Vanity Fair from 1935 containing the obituary of Lawrence of Arabia, a story on the new Bay Bridge in SF, the "Brown Bomber" Joe Louis, and pleas for peace in Europe. I would not pay anything for a PDF of this magazine. As a physical artifact, it was worth the $30 I paid for it. There is utility and value in physicality that cannot be found in digital ephemera.

Generally speaking, a Subaru has greater utility than a Ferrari, but no one complains that the Ferrari ought to be less expensive on that basis.

That said, I think it's an incorrect assertion that an e-book (with DRM) has greater utility than a physical book.

A Ferrari might get you laid. That's enough utility for some people.

part of the argument hinges on how fungible you (and consumers in general) perceive informational goods.

Another weird use of the star rating system: a lot of textbooks seem to get one-star reviews from people who really don't have a problem with the book but didn't like the instructor of the course they used it in.

Should the boxed version of Photoshop cost more than the digital download because the boxed version comes with higher marginal costs? What if additionally we made the assumption (which Ariely is making here) that the digital download version came with more features and conveniences? I think that our intuition (especially from the software maker's standpoint) is that they could at least charge the same for both.

According to classical economics, the manufacturer's costs should not have anything at all to do with deciding how much to pay for something. It's simply, "how much do I enjoy this?" Of course as we see consumers don't behave this way.

For Photoshop, the box is just a way of getting the product to you.

For a book, the physical object may be as much a part of the product as the information contained within.

It's not an apples-to-apples comparison.

To quote a reviewer: "We should not let the publishers hold us hostage because we prefer to read the electronic format.”

I think you're missing the principle matter: people are complaining that ebooks shouldn't cost as much as physical books because they cost less to produce. Nobody is saying they prefer physical books because they are physical objects, and nobody seems to be treating them like anything other than the same thing delivered in different ways at different costs. Find me evidence please that this whole thing isn't driven by the perception of fairness in what the product costs to produce.

Edit: Vote me down all you want, but lack of evidence is still lack of evidence. I guess when emotions run high people lose the need to make rational points and just downvote. Give me some justification. Argue with me.

Quick point of etiquette - not only would I not downvote based on disagreement, it's actually IMPOSSIBLE to do so when the comment in question is in reply to one's own comment. It's a HN feature to avoid the downvote becoming a kneejerk response to back-and-forth discussions.

You use one reviewer's statement that that person prefers an ebook. There's no further explanation as to why (s)he prefers that format, and no data behind it - just a single comment.

I expect that the primary issue is, in fact, what you say - fairness over cost of production. That does not logically lead to your assertion that the two are "same thing delivered in different ways."

Here's an apt analogy. Let's assume that customers preferred paperback books to hardback books, possibly because paperback books are lighter, take up less shelf space, and are easier to hold in one hand. They are, obviously, less expensive to make. It seems to me almost a given that these same customers complaining would be giving one-star reviews to the publisher for charging more for the paperback version. Under a utility model this wouldn't happen.

Absolute, complete utter bullshit. Functionality be damned--- such a pricing structure makes it impossible for people to get beyond the initial rip-off. Nor should they; this is the same approach that gave us 20.00 CDs with a single desired track, year after year after year. The single star review is the equivalent of the pitchfork and torch, perhaps it is time the publishers looked out the castle window!

Yes, I think you're absolutely right. What the publisher is doing, and Ariely agreeing with is known as Value Pricing: http://en.wikipedia.org/wiki/Value-based_pricing.

It's a great way to make sure you optimize your short term profits. And yes, your customer feels ripped off by it as prices as well they should as you have just siphoned off their economic value.

Bye bye customer relationship!

The most annoying thing is that Kindle books often have mistakes because of misses in the conversion (OCR). I don't think I have a single Kindle book that doesn't have at least a 3-4 typos (1's for I)...and there's always one that tends to really throw me off for a second.

It really annoys me to pay more and have additional mistakes in my book.

The article makes two contradictory points. First that the standard view that the utility of a product should determine the price is confounded by real-world human factors and second that consumers should ignore real-world human factors and accept prices based on utility. Seems to me that since economics attempts to model the transactions and interactions of humans engaged in various types of commerce that trying to change the subject being studied so the models work better is a little misguided.

It is strange. The part of a book that costs money is paying someone to come up with and type in the words. Printing it is cheap. Sending you an e-book is cheap. Those cost nothing compared to the labor that goes into making a good book.

Therefore, to me, it makes sense that a book would be $15 regardless of whether I get a dead tree in the mail or not. (This assumes that there is no DRM on the e-book. If I can't lend the book or share it with friends, then the e-book is not as good as the real book.)

Your basic facts are wrong. The cost of author royalties is about 15% of the total retail price.

Here's a good breakdown for you.

http://journal.bookfinder.com/2009/03/breakdown-of-book-cost...

Based on a list price of $27.95:

$3.55 - Pre-production

$2.83 - Printing

$2.00 - Marketing

$2.80 - Wholesaler

$4.19 - Author Royalties

$12.58 - Profit

Some costs other than printing also go away with ebooks. Pre-production becomes cheaper because the design and typography are typically discarded. Marketing gets cheaper when there are no books to mail or galleys to pre-produce for reviewers. Wholesale costs are gone, because distribution is streamlined to go directly from the publisher to Amazon, Apple, or whomever.

An $28 hardcover, sold exclusively as an ebook, could quite comfortably be sold for $15-$20 and still cover all the author, production, and marketing costs without losing any profit for the publisher.

When sold in conjunction with a hardback, as most ebooks are, the price can go even lower. Just as with paperbacks, the sunk costs are borne by the hardback. This is why a $28 hardcover becomes a $14 paperback - and why it could easily be a $10 ebook.

The "profit" section exists because there's a middleman, but let's forget about that for the sake of this argument. The cost of the stack of glued-together pages costs $3. Everything else pays for the production of the book in some way.

Charging the same for the e-book, which needs marketing, copyediting, graphics design, etc., etc., just like the dead-tree book, means the author (or someone) gets $3 more. In this way, it doesn't seem particularly evil to not give the end-user that $3 discount. Just give the author $3 more.

(As an author, the last thing I want is an App-Store-style race to the bottom. I want more than $0.30 per copy of book I sell, paper or otherwise.)

Clearly, you didn't read what I wrote.

If you're going to debate an issue, at least do the courtesy of considering the comment you're replying to.

Clearly, you didn't read what I wrote.

If you're going to reply "Clearly, you didn't read what I wrote," at least do the courtesy of considering the comment you're replying to.

After I broke down the costs very clearly, showing why there are more expenses than just the printing which are saved with ebooks - and the real-world issue of sunk costs which go into ebooks and paperbacks - you still insisted on claiming that the only savings would be the printing costs.

The only thing you seem to have read was the numbers and the word profit.

What, exactly, do you think you said that wasn't considered?

What I mean to say: OK, thanks for the breakdown. But the breakdown of current costs don't necessarily mean they are going to stay that way forever.

The one thing that's constant is that the author needs to be compensated for his work in creating the book. From there, he can just self-publish his word document or whatever. No marketing, no editors, no overhead. Now if you account for the cost of paper to print that book out, it ends up not being very much. Therefore, it's pointless to want an ebook for less than a printed book. You're paying for the author, not paying for the paper.

I'm not sure what Dan's point is. Customers know that digital versions cost publishers much less money. Pricing them higher is obviously going to make users feel disrespected and think "greed indeed", as a review stated.

And maybe it's because Dan and I are using different definitions of rationality, but I also don't see how the 1-star reviews are irrational. I think fairness is a big part of rationality, and the thought that "as readers we should pay a little less attention to minor differences in price" is shockingly short-sighted. Behind the 1-star reviews is a deep-seated and completely rational worry that if the higher pricing works, in the long term, sellers will be encouraged to bump prices up even more.

Edit: There are comments to the blog post about how the pricing makes sense from a "traditional economics" viewpoint. I don't know much about it, but if traditional economics doesn't take humanness (respect, fairness) into account then it's just a failure on the subject's part.

>>...humanness (respect, fairness)...

“Cheapness” is the word you are looking for.

Reply on news.ycombinator.com