Undercutting is a valid strategy in the real world, because products can be marketed to differentiate themselves from similar products. Features such as, different colors, "heavy duty", lighter construction, faster, etc... But there is no marketing in Eco, and everyone sells the same medium rug. There is no market segmentation which makes one medium rug worth more than another.
Price wars kill the servers because, in order to make prices ever cheaper, the producer must cut job opportunities out of the marketplace to become more efficient. For example, eating scavenged free foods eliminates buying foods from the chef skills, therefore cutting chef labor from participating in the economy. This is why undercutting is a failed strategy long term.
In a healthy economy, prices should be slowly rising. Falling prices is the sign of a recession. Long term drop in prices due to module efficiency, competition price wars, and reduction of labor costs, creates a depression.
To be clear, I'm not advocating that competitors shouldn't compete on price/location/selection. But predatory pricing comes into play once modules are factored into pricing calculations for a retail price. Ideally, for this simulation, the product producers should agree that the minimum price a product should be priced, is based off the inputs of the recipe. The addition of modules to improve efficiency is kept as profit, and not to be used to prevent market competition.