Kindle Review – Kindle Fire Review, Kindle 4 Review

One of the great mysteries of publishing is whether used book sales are good or bad for the book publishing industry.

Is the Used Books Market Good? For Whom?

Well, let’s dig in and find out.

How big is the Used Books Market? Is it growing?

What gets measured, gets managed – Peter Drucker

The Publishing Industry is, unsurprisingly, rather vague when it comes to used books. There are, however, a few scattered sources of information -

1. Book Industry Study Group’s study published in 2006. Write-ups by Mike Shatzkin and NY Times reveal these data points - 
   

   Used Book Sales were projected to be $589 million in 2004.
   2003 to 2004 growth of 25% in used book sales (when you exclude textbooks and education).
   Online used book sales were up 33%.

   Projection of annual growth rate (in dollar sales) of 25% for the used books market between 2004 and 2010 (1% for new books).
   Used book sales are up 11% (when including textbooks and education), new book sales are down 1.9%.

2. A Publishers Weekly article (by the way, PW is now up for sale) from September 2004 and a CNN feature on Alibris give us -
   

   Alibris CEO Mart Manley thinks used books will someday make up 15-20% of the market.

   Alibris Sales:
   2004: $60 to $70 million.
   2005: ?
   2006: ?
   2007: $114 million.
   2008: $134 million.
   That’s almost exactly 20% year over year growth.

   Abebooks Sales (Amazon bought Abebooks in 2008):
   2003: $100 million (in the US).
   2004: $140 million (in the US).

3. From a source now untraceable – 2002 used book sales of $530 million, and 2003 used book sales of $533 million. 
4. Book Hunter Press and Albert N. Greco (The Book Publishing Industry) chime in with their estimates -
   

   Albert Greco: ‘Empirical Data on used books has never been collected’. Estimates range from $500 million to $1.5 billion.

   BHP: Estimated used book sales in 2003 at $540 million, and estimates an additional $74 million from large used booksellers who refused to share figures.

   BHP: Some 80% of total used book sales are made by bricks-and-mortar stores that also sell online.

   BHP: In 2003, market share of used book stores online -
   Abebooks: 39.2%.
   Amazon: 17.3%.
   Alibris: 12.7%.
   eBay: 9%
   Dealer Websites: 8.6%.

If we take all those wonderful numbers and juggle them around we get a few key figures -

1. Used Book Sales of $589 million (the BISG figure) seems a pretty reasonable, conservative estimate. 
2. 2009 used book sales of $1.466 billion - assuming 20% year over year growth.
   

   Taking the lower of Alibris’ growth rate (20% year over year) and BISG’s predicted growth rate (25% year over year).  

   Amazon’s acquisition of Abebooks in August 2008 is another indicator that the used book market might be flourishing.

3. Used Books are over 10% of the Books Market.
   

   If you strip out Elhi and Higher Education from this Publishers Weekly 2008 book sales report (and assume sales don’t decline in 2009, which they probably will) you get 2009 book sales of $14.372 billion. $1.466 billion in used book sales is more than 10% of that.

Leaves us with a $1.5 billion used books market, that has 20% growth rate, and has a 10% share of the book market (and it might be more).

The Used Book Market is not neccessarily a good thing

The general defence of used books is -

1. It makes books affordable. 
2. New book buyers know they can sell their books. 
3. It increases number of books users can buy as they have to spend less.
4. Users get introduced to new authors.

All of these sound great. However, what are the actual facts?

The big study that is held up as proof that used books are good is one where Prof. Ghose of the Stern School of Business argues in favor of used books. A New York Times article by Prof. Hal Varian is based on this paper and is the holy grail of the ‘used books are great’ school of thought. The paper’s key conclusions -

1. Only 16% of used book sales directly cannibalized new book sales.  
2. 84% of used book sales would not have occurred without used books.  
3. The approximate $45 million in losses to publishers and authors due to sale of used books at Amazon is OK as society benefits (Amazon to the tune of $63.2 million, and readers to the tune of $67.6 million). 

Although this is an intelligent study (the 2.5 hrs spent on it were quite fun) it ought not to be used to defend used book sales. Even beyond the obvious fact that publishers and authors taking ‘losses for the benefit of society’ is rather ridiculous.

To understand why - let’s look at the assumptions the paper makes.

The Study 

1. The biggest assumption the paper makes is that it ignores intent - You have Alibris, Abebooks, used book stores and many other used book sellers. The paper studies just Amazon.com user behavior and assumes that intent of Amazon.com customers is an accurate gauge of the book reading population.  
2. The entire outcome of the paper (that only 16% of used book sales cannibalize new book sales) is based on modelling 42,000 unit sales of 393 books. 
3. The study was a snapshot in time i.e. Sept 2002 to Mar 3002, and April to July 2004. 

The Model

1. The model, as its limited to Amazon, doesn’t consider competition between retailers.
2. The model doesn’t consider flow within the used book market i.e. users reselling books.
3. The model uses a formula to map Amazon sales rank to actual sales figures. This is actually one of the quite brilliant parts and worth further study. It does, however, introduce a randomness factor.

Comparing Numbers from two completely different Estimation Models

The part that makes their ‘only 16% of sales are cannibalized’ argument invalid is –  

1. They use their study and their model to get a $4.63 million figure for sales cannibalized.  
2. They contrast that with a total used book sales figure they get by a simplistic guesstimate  -
   

   Take Amazon’s book sales.
   Take Amazon’s percentage of used goods sold (23%) and assume that applies to books too.
   Combine that to arrive at a figure for used book sales.

That means they’re contrasting two numbers derived from completely different estimation methods.

So …

What are the Hard Facts – What Impact do Used Book Sales have?

Unfortunately, there haven’t been any studies done apart from the one by BISG so we don’t really know. 

The only hard data points we have are -

From the Books Industry

1. Publishers and Authors get exactly 0% of used book sales.  
2. Even the Prof. Ghose study, while supporting used books, said that 16% of new book sales are cannibalized.
3. Used Book Sales are growing faster than New Book Sales.

From the Textbook Industry

1. In 2004, 30% of the textbook market was used books.  
2. The used book market kills textbook sales over time -
   

   Textbook Sell Through:
   1st year: 90%
   2nd year: 50%
   3rd year: 10%

Note that those figures are present even though used textbooks at college stores are 70-80% of the new price. 

The Used Books Conundrum - How could we ‘save’ New Book Sales? Should We?

There are some ways to eliminate the used books market hinted at in Prof. Ghose’s paper -

1. Increase prices of new books – In this economic environment this is ruled out. Plus there’s the danger that, like in textbooks, it just increases used book sales and piracy.
2. Restrict the used book market - Rather difficult thanks to the First Sale doctrine.   
3. Leasing as opposed to selling – This is exactly the direction eBooks are headed in. 
4. Rental Market – This is very similar to the above. It’s just replacing terminology and perhaps introducing different pricing models.

‘ Should We Save New Book Sales’ is a very personal question. However, it might not be up to us to decide.

Kindle and ebooks - the perfect solution to the Used Books Conundrum

If we were to design something that eliminated used books while still providing reasonable value to customers and authors/publishers we might look at -

1. Cheaper Prices for new books to compensate for zero resale value.  
2. Elimination of the Used Book Market by not allowing resales. 
3. An alternate method for promoting authors – perhaps free books, perhaps samples.
4. Perhaps even Eliminating sharing and giving away of books i.e. no more bookcrossing and no more sharing with friends.

Look at the model that Amazon has propagated and Google (with its ‘books in the cloud’) and B&N are copying -

Its almost the perfect solution to eliminate used books and ensure that used books don’t hurt new book sales.

It should not be a surprise that Amazon i.e. the company that has the most data and the best understanding of customer behavior (with both new and used books) is the one to implement a solid, working solution.

Peter Drucker is right – What gets measured, gets managed.

While book publishers have no idea of numbers (or won’t reveal they do) and textbook publishers are struggling, Amazon (thanks to having in-depth numbers) already has a working solution that completely eliminates the problem of used book sales.

When’s the last time you bought a used Kindle Book?

Filed under: kindle, publishing Tagged: | future of publishing, kindle vs used books, used books