It all started when I read this post from Carrie Rubin.  In it, she describes how people make money by entering book giveaways on sites like Goodreads, etc. and then immediately selling the free books they have won, without reviewing or even reading them.

I have an economics degree, so I began thinking about the incentives that cause this, and what adjustments you could make to the market to fix it. It made for a nice thought experiment. And, well, maybe it got a little out of hand.  But I’m posting it anyway; just for fun.

First of all, we need to discuss the concept of expected value. Wikipedia has a nice summary, using the game of roulette as an example, which I quote below:

Suppose random variable X represents the (monetary) outcome of a $1 bet on a single number (“straight up” bet). If the bet wins (which happens with probability 1/38 in American roulette), the payoff is $35; otherwise the player loses the bet. The expected profit from such a bet will be

E [gain from $1 bet] = -$1 x 37/38 + $35 x 1/38 = -$0.0526

i.e. the bet of $1 stands to lose $0.0526, so its expected value is -$0.0526.

Note that the expected value of a bet in a roulette game is negative.  This is why casinos make money and gamblers typically don’t.  The game is designed to be rigged against the player.

In the book giveaway scenario however, the “player” is not required to pay anything to enter.  The only cost to them is the opportunity cost of the time it takes to enter a giveaway, which is minimal once you have created an account.

Of course, to have a realistic shot at winning anything, you have to enter a lot of giveaways.  So the expected value is the sum of the value of each book times the probability of winning it.

Since there is no monetary cost to entering giveaways, the “player” only stands to gain by doing it.

The author, on the other hand, has little incentive to give the book away.  They only will benefit if the recipient likes the book and makes it known to others.  A bad review, or no review at all, goes down as a loss for the author. If the recipient then sells the book to someone else, it’s an even worse loss, because now multiple people are getting the benefit of the book without payment to the author.

How can we fix this?

One way would be to charge a fee to enter the contest, as in the case of the roulette example.  This would probably work too well–nobody would risk losing even $1 unless the potential reward were an extremely valuable book.  Hence, no one would enter the contests.

Another way would be to impose some limit on the number of giveaways a user is allowed to enter in a given timeframe.  After all, for it to be worth their while, the contestants must be entering a fairly high number of giveaways. Placing a cap on that could deter the book-scalpers.

But remember the original intent of the giveaways.  In an ideal world, the way it works is that the reader gets a free book, reads it, and reviews it.  Both the reader and the author benefit–in economics jargon, this means the outcome is “efficient”.

As part of my research for this post, I decided to find out if Goodreads takes reviews into account as part of the algorithm they use to pick winners.  According to this site, they do–citing the giveaway terms and conditions:

“If more people are interested in a book than there are copies available [which is nearly always the case], we will pick the winners at our discretion. The factors that go into our algorithm are: randomness, site activity, genre of books on your shelves, current phase of the moon, and more.”

I notice they say “site activity”, which is pretty vague, but I’ll assume it means that somehow or other they factor a user’s review track record into their chance of winning.

Tweaking that algorithm might go a long way towards fixing the problem.  But I don’t know what their algorithm is, so for the sake of this exercise, we’ll assume that it’s perfect.

Instead of changing the algorithm, another idea would be to change the terms of giveaways.  If a winner doesn’t post a review in some period of time (e.g. 30 days) they are required to pay a small fee–less than the price of a new copy of the book, but still enough to decrease the profitability of re-selling it.

I think this would increase the effectiveness of giveaways.  It incentivizes (to use that horrible word that only an economist could write) giving reviews, but while still benefiting the recipient, since even if they don’t review it, they are still getting the book at a cheaper rate.

That said, there are some potential problems with this plan:

1. What is the mechanism for charging people?  Goodreads does not require credit card information to make an account. (It’s much more straightforward on Amazon; there, all users have some sort of account that can be billed.)
2. It could lead to lots of garbage reviews.  People are likely to post short, unhelpful reviews to get their free copy.

We can still do better than this.

Remember, only physical books can be re-sold.  They can’t do much with a free eBook.  So, how about this: when someone wins the contest, they automatically get a free electronic version of the book, but to get the physical copy, they have to review it.

Amazon has a free reader app that runs on almost anything, so chances are, if you have a device that can be used to enter a GoodReads giveaway, it can also read an eBook.  And Amazon owns GoodReads, so it would be easy enough to set this system up between the two sites.

There’s still potentially a problem with an incentive to give garbage reviews, but it’s lessened considerably by the fact that the reader gets rewarded for posting a review, rather than punished for not posting a review.

What do you say, readers? Could this work?  Is it totally insane?  Do I think too much about weird stuff? Is this why everyone hates economists? Let me know what you think!