It is the title of a new USC white paper with subtitles of Daren Filson, Karen Van Nuys, Darius Lakdawalla and Dana Goldman.How much does your profits lead to new drug development?‘
What is the elasticity of innovation?
In general, due to the percentage change of expected sales, we measure the rate of change of innovation using new drug approval or starting steps 1, 2 or 3.
Future profit.
In fact, it is an important change in profits, but the future sales are much more observed and predictable than the future interests. Therefore, the author focuses on the elasticity of innovation regarding profits rather than profit.
How much will revenue affects the possibility of new drug development?
All studies show that elasticity is positive (eg, the lower the profit, the less R & D, but the estimates are greatly different, but we have a typical long -range elasticity related to US sales within 0.25 to 1.5, and you can expect a 2.5% to 15% pharmaceutical innovation whenever the expected profit decreases by 10% I insist.
What leads the volatility of this estimates?
One important question is why is there a big range in these estimates? Certainly other research designs are important (see below). The authors also argue that factors such as “time horizon, price fluctuations, drug development costs, barriers and other market factors for value -based prices” affect the size of innovation.
What methodology is used in the literature to estimate the elasticity of innovation?
- Cross -sectionalTo estimate the elasticity, use a change in profits across the treatment class (or other analysis units). For example, they can infer elasticity by comparing “high profits” and “low -level” classes (eg, for example:: LichTenberg (2005) and Civan and Maloney (2009)).
- Aggregate: Over time, use a change in industrial -level revenue (e.g.:: Giaccotto, Santerre and Vernon (2005)))
- Panel data access: Includes the drug class “fixed effect” and includes the difference and continuous difference that is difficult to measure the class characteristics. In essence, this approach focuses on changes in profits in classes as a driver of innovation change in class. Such analysis usually needs to use “natural experiments” where income is differentiated in other sectors in the market. Examples of natural experiments include future demographic changes in the future or the appearance of Medicare Part D (e.g.:::: Acemoglu and Linn (2004); Dubois et al. (2015); Blume-Kohout and Sood (2013)))
- Parameterized calculation model (aka rescue model)If you specify the target function, strategy set, and functions of the company in the business environment and include multiple companies in the model, the model must be in equilibrium. The parameters are chosen to match the actual world parameters (eg average R & D expenses) and corrected the model output to match the actual results (eg, the average flow of the New Testament). (yes : Abbott and Vernon (2007); Filson (2012); Adams (2021)))
The author argues that the panel approach and parameters are preferred.
In the case of a study on preferred panel or calculation approach, what innovation estimates did you reach?
The author has a nice table that summarizes the discovery below.
Great work of USC colleagues! I recommend you read the entire article here.
