Music piracy: Bad for record sales but good for the iPod?

- I construct a unique conjoint survey data on music piracy and iPod consumption.
- I use a Bayesian method to estimate the joint demand for music and iPods.
- Music piracy decreases music sales by 24% to 42%.
- Music piracy contributes 12% to iPod sales among the students in the sample.
- Apple's increase in profits can compensate for the loss of musicians if music were free.

Music piracy is a double-edged sword for the music industry. On the one hand, it hurts record sales. On the other hand, it increases sales of its complements. To quantify the effect of music piracy, I construct a unique survey data set and use a Bayesian method to estimate the demand for music and iPods, and find three things. First, music piracy decreases music sales by 24% to 42%. Second, music piracy contributes 12% to iPod sales. Finally, counterfactual experiments show that, if music were free, the increase in Apple's profits from iPod can more than compensate the loss of musicians. The last result implies that a Pareto improving iPod tax is possible.