2025-06-26 ### 📜 Summary: The Rationality of Yedioth's Inventory Problem ![[yedioth 2025-06-26-11.svg]] %%[[yedioth 2025-06-26-11|🖋 Edit in Excalidraw]]%% using [pooling methods visualization cld](https://claude.ai/chat/53f2f444-1083-4d43-ad78-351b4fe362f1), This case study is a practical application of using rational, statistical tools to overcome costly, ingrained business biases. - company mathematically ensures that the leftover inventory (Q−D) will be a large fraction of the initial quantity, (by unconsciously? setting high C_u) | Section | Key Message Summary | | | --------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | **1. The Problem: A Conflict Between Passion and Profit** | Yedioth's goal is to be profitable, but its actions are driven by a powerful "passion": the fear of losing a single sale. <br><br>This leads to the irrational practice of systematic overproduction, where the high, certain costs of returned magazines (25-35% of stock) are not rationally weighed against the uncertain benefit of preventing a stockout. <br><br>The company is trapped in a cognitive bias, prioritizing the prevention of a visible loss (an empty shelf) over the acceptance of a less visible but larger one (scrapped inventory). | | | **2. The Solution: Statistical Rationality as a Tool** | The case introduces the Newsvendor model as the instrument of rationality. It is a tool for making an optimal decision under uncertainty by using knowledge (historical sales data) to model demand as a probability distribution. <br><br>This allows Yedioth to move from a decision based on fear to one based on a calculated trade-off between the cost of ordering too much and the cost of ordering too little. | | | **3. The Application: The Power of Collective Rationality (Pooling)** | The solution explores three ways to apply this rational tool, demonstrating that greater information sharing and centralization lead to better outcomes. | | | | a. No Pooling: | This is the baseline state where each of the 8,000 retailers is treated independently. While rational at the micro-level, it is collectively inefficient, requiring the highest total inventory to meet the service goal. | | | **b. Pooling by Agent:** This represents a more rational choice. By grouping retailers by sales agent, the system leverages risk pooling—the demand variability of the group is less than the sum of the individual variabilities6666. This is a practical step toward collective rationality that yields significant inventory reduction7. | | | | c. Full Pooling: | This is the theoretical ideal of perfect rationality, where all demand is aggregated into a single pool.<br>It demonstrates the maximum possible benefit of centralization and shows that a perfectly rational system requires the least inventory to achieve the same goal, drastically reducing waste. | | **4. The Conclusion: Rationality Creates Value** | The key lesson is that operational models are not just mathematical exercises; they are tools that enforce rationality on complex business systems. By compelling a data-driven approach, the Newsvendor model helps Yedioth overcome its institutional biases. The tangible benefit is a dramatic reduction in wasted inventory and cost, proving that a more rational process is a more profitable one. This exemplifies Pinker's core argument: rationality is the application of knowledge to achieve one's goals. | | - **The Cost of Overage (Co​):** The loss incurred for each magazine that is produced but not sold. For Yedioth, this includes the production cost plus the additional costs of collecting and scrapping the returned magazine, as they offer a full refund to the retailer1. - **The Cost of Underage (Cu​):** The opportunity cost for each unit of unmet demand, which is the profit lost from a sale that could not be made. Yedioth perceives this cost to be very high due to the potential loss of readers and advertising "eyeballs"2. A rational decision-maker calculates the optimal service level (the probability of meeting all demand) using the critical ratio. The optimal order quantity Q∗ is set where the probability of selling one more unit is no longer justified by the potential profit. This is defined as: P(Demand≤Q∗)=Cu​+Co​Cu​​ Yedioth's behavior demonstrates a severe miscalculation of this ratio. 1. **Implicit Inflation of Cu​**: Their instinct to "over-produce to make sure no potential sales were lost" 3 means they act as if the cost of a single lost sale (Cu​) is astronomically high. 2. **Implicit Dismissal of Co​**: They simultaneously act as if the cost of overage (Co​) is negligible, despite evidence that returns are between 25% and 35% and incur real reverse supply chain costs444. By implicitly setting Cu​>>Co​, their required critical ratio Cu​+Co​Cu​​ approaches 1. To satisfy this, they must choose an order quantity Q far into the upper tail of the demand distribution. This action mathematically guarantees a high volume of leftover inventory (overage), perfectly explaining the observed 25-35% return rate. Their practice is irrational because a formal calculation would almost certainly not justify such an extreme service level, revealing that the certain cost of overproduction outweighs the uncertain benefit of preventing the last potential stockout.