Rent or Buy?

Economics, Movies

Last year I presented a paper co-authored with my colleague David Lang at CSU-Sacramento entitled “Does Sex Sell?” It looks at the determinants of box office revenues, using a pretty large sample of movies released between 1993 and 2004. In it, we also look at foreign box office and discuss the importance of DVD sales, but at the time we did not have the data on the DVDs. Now we have it, thanks to, and we are writing another paper to include this. In the process, I came across what I think is some interesting information, and I thought I’d share.

The table below shows the top 20 DVDs purchased for 2008 as well as the top 20 DVDs rented in 2008. Movies in bold on the rental list also appear on the purchases list. The value in the “Rent vs. Buy” column for those movies shows the difference in rank between the rental list and the purchase list.


I think it’s interesting to note that only 7 of the movies are on both lists. For every movie but Iron Man, the movie ranks higher on the rental list than on the sales list. Apparently most people who wanted to watch Iron Man bought the DVD, but the rental market was still strong. For the rest of those movies, they weren’t good enough to buy, but people still wanted to see them.

Here’s what I see when I look at these two lists. I see that there are two kinds of movies that people rent: Academy Award Nominees and crap. People wanted to watch Michael Clayton (good movie, btw) and No Country For Old Men (boring, IMO) so they could have some idea of which should win the Oscar. And people wanted to watch Good Luck Chuck (for some unknown reason) but knew based on its horrible box office performance and reviews that they would probably only want to watch it once — and even then probably only for $1 at Redbox. Baby Mama and Fool’s Gold are in this same league too.

I think it’s noteworthy that The Dark Knight was the best-selling movie of the year at the box office, as well as the best-selling DVD, but it is nowhere on the rental list. I guess when everybody has already seen the movie and everybody knows someone who owns the DVD, there’s no reason to rent it.

Of the top 20 movies sold, 9 of them can legitimately be called kid’s movies. (And yes, it can be a kid’s movie if it has The Rock in it. He’s the new Ice Cube — Ice Cube after he sold out, that is, and did Are We There Yet? and Are We Done Yet? Now The Rock does Race to Witch Mountain and Game Plan. I smell what the Rock is cookin’ and it doesn’t smell so good.) Only one of these 9 movies, Game Plan, is also on the rental list. Parents buy their kids the DVD, plain and simple.

Part of what we’re looking at in the DVD paper is how different aspects of movies affect DVD sales and how that differs from box office revenues. We have already found, for example, that a movie’s sexual content does not have a positive impact on box office revenues for R-rated movies. However, it does have a positive impact on DVD sales for R-rated movies. Conclusion: people want to watch those movies in the privacy of their own homes.

Some facts on the DVD/VHS market in case you’re interested: The DVD sales market is by far the most profitable distribution vehicle for movie studios. The revenues from DVD sales are more than the worldwide box office of movies, and the studios get a larger chunk of the revenues on DVDs (about half of box office revenue goes to your theater, the other half goes to the studios who made the movie; with DVDs, studios get more like 80% of the revenue). The DVD rental market shrunk last year, and only makes up about 1/5 of the revenue of the DVD sales market. That market shrunk last year too, as the economy took a turn for the worse. Blu-ray sales picked up, but not enough to offset the DVD sales decrease. VHS is practically dead: VHS rentals make up less than 4% of all rentals.

Netflix sent me an e-mail this week saying that if I wanted to stay on my same plan (3 discs at a time, including Blu-ray), I would have to pay an extra $3 per month. Until now, they were charging me an extra $1 for Blu-ray, and they are increasing that charge to $4. I promptly changed my plan to 2 discs at a time and am saving $4/month. One might think that people who can afford a Blu-ray player wouldn’t have a problem with an extra $3 a month, and I think that’s what Netflix is counting on. But I’m cheaper than most people, so my demand for everything is more elastic. And rumor has it that Redbox is carrying Blu-ray in other places around the country for the same price of $1/day, and as long as that hits St. Cloud soon, I should be just fine.



  1. ProfSwitzer  •  Apr 2, 2009 @7:04 pm

    I received this from a friend on facebook and am reprinting here and answering. Her comment:

    I’m curious as to why you find it interesting that the Dark Knight is nowhere to be found on the rental list when it banked big time while in the theaters. I assume that everybody who was interested in seeing it saw it at the theater, therefore, completely bypassing the rental stage and going straight to buying the dvd. I guess where I’m interested is why it doesn’t happen that way more often. But then you have to take a look at 1) how big of hype batman was, esp w/ Heath Ledger’s death and 2) how entertaining of a movie it was compared to what else gets released. I dunno…love to hear your thoughts.

    My response:

    I did originally write that I found it interesting, but after that comment, I updated the post and changed it to “noteworthy.” But it got me thinking, and I’m not sure why it doesn’t happen more often, or even if it does. So I looked at the top box office numbers of 2008 for comparison. Here they are:

    1. The Dark Knight
    2. Iron Man
    3. Indiana Jones
    4. Hancock
    5. WALL-E
    6. Kung Fu Panda
    7. Madagascar 2
    8. Twilight
    9. Quantum of Solace
    10. Horton Hears a Who
    11. Sex and the City
    12. Mamma Mia! (I feel so lame typing the !)
    13. Juno
    14. Prince Caspian
    15. Incredible Hulk

    Most of these movies are also on the sales list (some, like Twilight and Quantum of Solace, did not have the DVD released until 2009 so they aren’t on the sales list). I guess the reason I wrote that it was interesting is because we’re looking at box office sales determinants and DVD sales determinants, and they’re not always the same, but there are some similarities. I just noticed such a stark difference between what people rent and what people buy, and the fact that The Dark Knight didn’t show up anywhere on the rental list caught my eye. Iron Man is second in box office and second in DVD sales, but it’s also on the rental list.

    But the main point presented in the comment above, I think, is a two-part question I would love to answer: how well does a movie have to do in theaters for people to want to rent it, and at what point does it become a movie that is so good people have to buy it to keep in their collection? If only they provided actual rental numbers, we could write that paper too…

  2. Zach  •  Apr 5, 2009 @1:32 am

    Prof Switzer,
    Have you thought about talking with Netflix or possibly Redbox and seeing if you could get there numbers, in exchange for you doing work for them? I find these “new” DVD numbers quite interesting personally because that is what my Seminar Paper is on. I might have to incorporate that into my final model if it is easy to get into excel or e-views.
    Another question that I have for you on this, is how reliable do you think these numbers are? I only ask because some of the other numbers on the website look a little off.

  3. ProfSwitzer  •  Apr 5, 2009 @11:01 am

    There are some slight variations depending on the source, possibly because DVD numbers continue to trickle in over time and perhaps different sources update their numbers with different frequency. The rankings of DVD sales and rentals shown in the post are from Variety, and they differ slightly from the numbers at I know that the-numbers breaks it up by sales year, so if The Dark Knight is released in 2008 and is still bought in 2009, it shows up in two different years; the 2008 numbers would be set in stone and the 2009 numbers would continually change as more and more people continue to buy the movie, even to this date. Part of the problem with using DVD sales or rental data is that they are always changing, so you either need to impose a time constraint (sales within the first 6 months) or just understand that there will always be some slight differences. The administrator for the-numbers says they’re confident their numbers are +/- 5%, and they reconcile their results with numbers released by studios when they come out eventually. So for new movies, they are probably a little less reliable, but for movies released years ago, they’re probably really close. Even when there is a difference between web sites, I find it’s usually less than 1%. So in terms of modeling this and testing what determines sales, I think the results would be very much the same regardless of whose numbers you use.

    I haven’t thought of contacting Redbox of Netflix. Usually with small companies they will help you with data if you give them results and help them answer questions. I guess I must just be assuming that those companies already have people working on those things, so why would they need me? But it’s worth a shot…thanks for the suggestion. I have nothing to lose by asking.

1 Trackback

Leave a Reply