Theory 01
Information as an Economic Good
Information differs from physical goods in fundamental ways. Shapiro & Varian (1998) highlight that information is expensive to produce but cheap to reproduce (high fixed costs, low marginal costs).
Non-Rivalrous
Consumption by one person doesn't prevent consumption by another. Thousands can use the same dataset simultaneously.
Non-Excludable
Hard to stop others from using it once it's out (unless protected by IP or encryption).
Experience Good
You don't know the value of information until you have consumed it. (e.g., a movie or a research report).
Zero Marginal Cost
The 1,000,000th copy of a digital file costs virtually nothing to create.
Theory 02
The Information Value Chain
Data by itself has no value. Value is created through a transformation process: Data → Information → Knowledge → Wisdom/Action.
Theory 03
Scale, Scope & Returns
Information economies exhibit Increasing Returns. Iansiti & Lakhani (2020) argue that digital operating models allow firms to grow in Scale and Scope without the traditional diminishing returns of human-centric organizations.
Key Definitions
- Scale: Ability to serve more customers with minimal cost increase. (Fixed costs are spread thin).
- Scope: Ability to enter new business lines by reusing existing data/assets.
- Learning: The more the system is used, the more data it generates, making the product better for everyone.
Case Application
Novo Nordisk — Information Arbitrage?
How information economics applies to the pharmaceutical case:
| Concept | Application |
|---|---|
| High Fixed Cost | Pharma has massive R&D costs (production of info). Distribution is cheap once the formula is known. |
| Scope Economies | Data from one obesity trial can potentially inform research in related cardiovascular or kidney diseases. |
| Experience Good | Clinical trial results are only "known" after the study is complete — hence the high risk of drug development. |
Exam Preparation
Likely Oral Exam Questions
-
Core Why is information described as "expensive to produce but cheap to reproduce"? ▶
- First-copy costs: Most of the cost is in the initial creation (R&D, writing code, clinical trials).
- Marginal costs: Digital reproduction (copying a file) is nearly zero.
- Implication: This creates massive economies of scale and encourages "Winner-Take-All" market dynamics.
-
Core How does the DIKW framework help explain the value of AI in firms? ▶
- Firms have lots of **Data** (unstructured clinical logs).
- AI transforms this into **Information** (patterns of drug response).
- This builds organizational **Knowledge** (understanding biological pathways).
- Ultimately, it enables faster/better **Action** (launching the right drug trial).
-
Synthesis What are the risks of "Increasing Returns" for an incumbent like Novo Nordisk? ▶
- Increasing returns favor the leader. If a competitor builds a better "AI Learning Loop", they may become unstoppable.
- Novo must use its current data advantage to secure its position before a more agile digital-first competitor scales.