How would you use a randomized controlled trial (see box in Chapter 2) to assess the importance of “quality-quantity tradeoffs” in determining how much parents invest in their children? Can you think of a “natural experiment” that allows for similar inference without so many ethical problems?