Nonparametric Demand Identification Without Exogenous Characteristics
Analysis
This paper challenges the conventional wisdom that exogenous product characteristics are necessary for identifying differentiated product demand. It proposes a method using 'recentered instruments' that combines price shocks and endogenous characteristics, offering a potentially more flexible approach. The core contribution lies in demonstrating identification under weaker assumptions and introducing the 'faithfulness' condition, which is argued to be a technical, rather than economic, restriction. This could have significant implications for empirical work in industrial organization, allowing researchers to identify demand functions in situations where exogenous characteristic data is unavailable or unreliable.
Key Takeaways
- •Challenges the necessity of exogenous product characteristics for demand identification.
- •Proposes 'recentered instruments' as an alternative method.
- •Introduces the 'faithfulness' condition.
- •Offers a potentially more flexible approach to demand estimation.
“Price counterfactuals are nonparametrically identified by recentered instruments -- which combine exogenous shocks to prices with endogenous product characteristics -- under a weaker index restriction and a new condition we term faithfulness.”