In light of a new Parliamentary inquiry over price gouging for software products, many analysts, including both Stephen King and Joshua Gans here at Economics.com.au, offered explanations for why prices for software in Australia are higher than, say, in the US.
Stephen observes that higher prices are the result of price discrimination provided Australian consumers are less price sensitive than their US counterparts. To explain the lower price sensitivity of consumers, he suggests that Australians have fewer opportunities to purchase abroad; but this obviously applies less to software purchases than to other products. Joshua points out that if digital piracy is more prevalent in Australia, then the price sensitive segment of consumers does not participate in the official market and the remaining consumers are less price sensitive. More often, “behavioural” or ad hoc explanations are invoked for why Australians are less price sensitive. Stephen concludes with the open question:
Why is Australian demand less sensitive to price may be unanswered, but it appears that the relevant software sellers believe it is true and so charge us more.
A possible answer for why this price discrimination occurs lies in the Australian market structure beyond the software market. Consider for a moment a world with two commodities, “real goods” and “software”, which are substitutes: if the price for software rises, consumers buy less software and more real goods, and vice versa. Now if the software provider is a monopolist, then its price and the price for real goods are strategic complements: an increase of the price for real goods raises the price the software publisher should charge for its software; a lower price for real goods lowers the optimal price with which the monopolist responds. When both types of goods are sold by monopolists of this sort then the resulting price equilibrium is called Bertrand equilibrium with differentiated products and it results in prices above the perfectly competitive level and below the levels the monopolies would charge if the two markets did not interact.
Suppose now that the market for real goods in the US is competitive while in Australia this market is dominated by a small number of firms, resulting in prices for real goods (those other than software) that are higher in Australia than in the US. In the absence of software arbitrage the software monopolist will now optimally charge a lower price in the US than in Australia—simply by responding to the lower price for real goods in the US. The software publisher thus benefits from the lack of competition in the market for real goods in Australia and enjoys higher profits.
What should be done?
Since the market structure in the market for real goods affects the pricing decision by the software provider, increasing the competitiveness of the market for real goods would not only improve prices for real goods but also discipline the software publisher. Thus, the focus of the Parliamentary inquiry and of the Productivity Commission should not be on software providers specifically but rather on the retail sector as a whole. The software providers are simply responding to the imperfections of the larger market.
Suppose that instead of improving the competitiveness of the real goods market the regulator enacted a “non-discrimination law”, requiring a software company that sells software in Australia to allow Australian consumers to also purchase off its international portals. Facilitating this software arbitrage might(*) induce software publishers to reduce Australian prices closer to their international levels. A lower software price domestically would also have a spill-over effect on the domestic markets for real goods since on average the price for real goods is a strategic complement to software prices as well. Realistically this effect is likely to be small considering that the real-goods market is a composite of many heterogeneous markets and, as a whole, is much larger than the market for software.
For some of the real goods, though, the effect would be just the opposite. For goods that are complements to software—such as computers—a lower software price would provide an incentive for the seller to charge a higher price.
These latter observations are loosely related to the theory of the second best, formulated by the Australian economist Kelvin Lancaster and Richard Lipsey, which holds that if several markets suffer from imperfections, correcting one of them and moving it closer to its competitive optimum will not necessarily be welfare improving.
(*) I say “might” here since, in fact, it might not—but that is beside the point.