The Market for Digital Goods An Overview. Information Commodities Economic characterization of commodities –A good or service is completely characterized.

The Market for Digital GoodsAn Overview

Information Commodities

• Economic characterization of commodities– A good or service is completely characterized by its

physical description, the place in which it is available, the date at which it is available, and any contingencies under which it is available.

– Like other commodities, digital commodities can be durable (in which case we call them goods) or perishable (in which case we call them services).

• Digital goods typically exist in code (software)

• Digital services typically involve processing information without the process itself having any permanent existence.

Characteristics of Information Products

• Physical characteristics– Indestructibility

• Coase problem• Obsolescence issues: What is a used information good?

– Transmutability• Intellectual property right issues

– Reproducibility• Excludability• Pricing: Product differentiation and market segmentation• Property right issues

Characteristics of Information Products

• Spatial and Temporal Characteristics– The internet is everywhere and it never sleeps

• Enhancement to competition– Arbitrage opportunities across different market locations or across

different time periods.

– Global markets available to all potential sellers or buyers

– Applications of technology to enhance competition: auctions, search.

• Breaking of local market power– Information allows replacement of distribution networks with just in

time delivery systems.

Characteristics of Information Products

• Contingent characteristics– Real-time, or near real-time, interactivity can reduce

the need to anticipate (and write contracts contingent on) unforeseen random events.

– Networked interactions themselves introduce new random events that can increase or decrease the value of a digital product (new sources of uncertainty associated with technology, new problems of information asymmetries leading to problems of moral hazard or adverse selection.

– Real-time financial transactions also significantly expand the scope for contingent trades.

Markets for Information

• Complements or Substitutes?– Information generated in any transaction will have

value, at least to the parties to the transaction. This information is complementary to the transaction, and can itself be viewed as a commodity with economic value.

– Information and related digital goods having value in and of themselves will compete with other substitute goods or services in the market directly, rather than as an adjunct good or service to some other transaction.

Markets for Information, cont.

• We focus on both aspects of the market for information, beginning with an analysis of markets for pure digital goods or services, then turning to adjacent markets.

Strategic Issues in Markets for Digital Goods

• Indestructibility– Innovation or Standardization?

• “If you don’t cannibalize your own market, someone else will.”

– Related issues of pricing and market organization• Cost structure

• Market segmentation

Strategic Issues in Markets for Digital Goods

• Reproducibility– Can you even make a market?

• Excluding non-payers• Copy protection versus customer annoyance• Pricing in the face of easy reproducibility

– Copyright and Intellectual Property• How to keep rivals from mimicking your success• How to ensure that added value gets compensated

Pricing Digital Goods

Microsoft vs. Britannica

• Pre-internet: 32 volume set of Encyclopedia Britannica cost $1600 in hardback.

• Microsoft’s strategy for marketing electronic encyclopedic services:– purchase Funk & Wagnalls Encyclopedia

– use the content of F&W to produce a CD with multimedia enhancements and a user friendly search facility

– market the result as MS Encarta for $49.95

Microsoft vs. Britannica

• Result:– Britannica loses significant market share to the search

features and multimedia enhancements of Encarta and other electronic encyclopedia’s.

– Britannica fights back:• Online version for libraries (cost: $2000)

– But households, smaller schools and libraries continue to defect to cheaper electronic encyclopedias and online encyclopedia services

• Britannica offers an online subscription for individuals for $120/yr and CDROM for $200.

– Households still not willing to pay 4 times the cost of Encarta

Microsoft vs. Britannica

– Shake out• Jacob Safra purchases Britannica’s parent company,

disbands sales network, and begins aggressive price-cutting.

• Britannica’s CDROM version currently sells for $49.95 after mail-in rebate

The Coase Experiment

• Analysis of the data

Round 1: Monopoly Seller (73-469)

0

500

1000

1500

1 4 7

10

13

16

19

22

25

28

31

Transaction

Pri

ce

The Coase Experiment

Round 1 (45-857)

0

1000

2000

3000

4000

1 2 3 4 5 6 7 8 9

Transaction

Pri

ce

The Coase Experiment

Round 2 (73-469)

0

200

400

600

800

1 2 3 4 5 6 7 8 9 10 11 12

Transaction

Pri

ce

The Coase Experiment

Round 2 (45-857)

0

1000

2000

30001 4 7

10

13

16

19

22

25

28

31

34

Transaction

Pri

ce

The Coase Experiment

Round 3 (73-469)

0

500

1000

1500

0 2 4 6 8 10

Quantity Sold

Pric

e

The Coase Experiment

Round 3 (45-857)

0

1000

2000

3000

40001 3 5 7 9

11

13

15

17

19

21

23

25

Transaction

Pri

ce

Experiment

• Supply and Demand Analysis

• Conclusion: With even a small amount of competition, prices are driven down to marginal cost.

The Cost of Information

• The cost of producing the first unit of a digital good is generally not small, and can be substantial.

• As we have seen, the indestructibility and reproducibility of digital goods means that the marginal cost of producing an additional unit of the good is close to zero.

• Because the cost of storing and transmitting stored information is cheap (and continues to get cheaper), there are also no effective capacity constraints on the production of digital goods.

Cost Diagram

Implications

• Globally declining average costs imply significant economies of scale.

• Minimum efficient scale can be on the order of the whole market

• We should not expect the see highly competitive market structures.

Implications

• What market structures should be expect to see?– Markets with a dominant firm

• Microsoft comes to mind

– Differentiated Product Markets

– Commoditized information markets• Digital goods selling at marginal cost

• Free information products (maps, telephone information, email addresses, news, stock price quotes, etc.)

• We examine models of each of these structures in turn.