The article compares different economic models to see which one best fits real-world data on consumption and income. By using a Bayesian approach, the researchers found that a model with both risk aversion and discount factor heterogeneity is unlikely to have generated the data. They also discovered that the values of the unemployment rate during booms and expansions greatly impact the relationship between consumption and income. This suggests that policymakers should be cautious when using certain economic models to make decisions about the economy.