4.2 Market Size in Innovation: Theory and Evidence From the Pharmaceutical Industry

Podcast

Play an AI-generated podcast conversation about this lesson

Questions and Answers

According to the provided text, what is the relationship between market size and entry for generic drugs?

Generic drug entry is equally responsive to changes in market size as non-generic drug entry.

Generic drug entry is less responsive to changes in market size than non-generic drug entry.

Generic drug entry is not responsive to changes in market size.

Generic drug entry is more responsive to changes in market size than non-generic drug entry. (correct)

In the context of the model, what is used as a proxy for market size, mj(t)?

The number of clinical trials for drug category c at time t.

The number of patents for new drugs in category c.

Actual market size data for drug line j.

Potential market size driven by demographic changes, Mct. (correct)

What does the provided equation log nj(t) = constant + log δj + log mj(t) represent?

The relationship between the number of clinical trials and drug approvals.

The relationship between generic and non-generic drug market sizes.

The relationship between new drug entry and market size when r approaches zero. (correct)

The relationship between patents and drug innovation rates.

Which of the following is NOT a direct proxy for innovation rates?

Potential market size Signup and view all the answers

What is a key distinction the study intends to investigate regarding the relationship between market size and drug entry?

Whether there's a difference in the relationship for generic versus non-generic drugs. Signup and view all the answers

Approximately how many non-generic drugs were approved compared to generic drugs in the dataset from 1970-2000?

About half as many non-generics as generics. Signup and view all the answers

How were the 34 drug approval categories grouped for analysis?

By the age group that accounted for the largest fraction of use in that category. Signup and view all the answers

What trend was observed when comparing the income share and drug approvals for the 0-20 age group?

Both income share and drug approvals showed a downward trend. Signup and view all the answers

What is the stated reason for fluctuations in the total number of drug approvals?

A number of institutional changes, including discovery of corrupt FDA officials. Signup and view all the answers

What happened to the drug approval process for generics in the early 1990s?

It was greatly slowed down. Signup and view all the answers

What was the implication of finding positive association between contemporaneous changes in population share and changes in drug approvals for corresponding age groups?

Drug approvals are linked to the corresponding segment of the population. Signup and view all the answers

What statistical methods were used for analysis, according to the passage?

Non-linear least-squares (NLLS) and ordinary least-squares (OLS). Signup and view all the answers

What does the text suggest about the drug entry and income shares for the 30-50 age group?

Both the shares of income and drug entry increased. Signup and view all the answers

According to the model, what condition must be met for steady-state equilibrium regarding R&D level?

The R&D level must be constant, i.e., $ż_j(t) = 0$. Signup and view all the answers

In the steady state, which of the following factors would lead to an increase in R&D for a given drug line?

A decrease in interest rates Signup and view all the answers

In this extended model, how does the equilibrium behavior of R&D for a specific drug line relate to other drug lines?

It is completely independent of research and profitability in other drug lines. Signup and view all the answers

What conclusion can be drawn about the dynamics of $z_j(t)$ based on the provided analysis?

Starting away from the steady-state, the equilibrium $z_j(t)$ will immediately jump to its steady-state value. Signup and view all the answers

Consider a scenario where a future increase in market size $Y_j$ is announced at time $t_0$ for a future date $t̂ > t_0$. According to the model, what happens to $z_j(t)$ at $t_0$?

$z_j(t)$ will smoothly increases towards the new steady-state equilibrium starting at $t_0$. Signup and view all the answers

If a future market size is expected to increase, what occurs with the value $V_j(t|q_j)$ before the change in market size is realized?

$V̇_j(t|q_j)$ becomes less than 0. Signup and view all the answers

What is the implication of the fact that the right hand side of equation (15) is strictly increasing in $z_j(t)$ at $z_j(t)=z^S_j$?

It implies that there can be at most one intersection with the 0 axis, and at this intersection point, $ż_j(t)/z_j(t)$ is increasing in $z_j(t)$. Signup and view all the answers

In the provided model, what does the term $M_{ct}$ represent?

The potential market size for category c at time t. Signup and view all the answers

In the given model when it is said that there are no transitional dynamics, what does this imply regarding R&D?

The R&D level immediately jumps to its new steady-state value when there is a change. Signup and view all the answers

What is the primary reason for using the logarithm of $N_{ct}$ as the dependent variable?

To ensure that drug category and time effects have proportional impacts. Signup and view all the answers

Why is the original model (equation 23) problematic when $N_{ct}$ equals 0?

The logarithm of $N_{ct}$ would be undefined, making estimation impossible. Signup and view all the answers

In equation (24), how is $Ñ_{ct}$ transformed for cases where $N_{ct}$ = 0?

$Ñ_{ct}$ is set to 1 when $N_{ct}$ = 0. Signup and view all the answers

What is the potential drawback of introducing the dummy variable $d_{ct}$ in equation (24)?

It introduces statistical bias because $d_{ct}$ is a function of $N_{ct}$. Signup and view all the answers

According to the provided equations, what is the impact on $z_j(t)$ if $\delta_j$ increases, assuming that Assumption (19) holds?

$z_j(t)$ decreases Signup and view all the answers

Based on the model, what happens to R&D, represented by $z_j(t)$, if Assumption (19) is not valid?

R&D, $z_j(t)$, becomes zero. Signup and view all the answers

According to the model, the entry rate of non-generics, $n_j(t)$, is influenced by which of the following factors?

Both the parameters $\mu$ and $\theta_j$ Signup and view all the answers

What does the model predict about the relationship between market size and the entry rates of both non-generics, $n_j(t)$, and generics, $g_j(t)$?

Both entry rates respond positively to market size Signup and view all the answers

In the context of the model, if $\mu = 0$, what is the state of the equilibrium?

The equilibrium is that of Proposition 1 Signup and view all the answers

According to the model, what factor determines whether the entry of generics or non-generics will respond more to changes in market size, $Y_j(t)$?

Whether $\mu$ is larger than $(\delta_j - \mu \theta_j) / (\theta_j - \delta_j)$ Signup and view all the answers

What happens to the limiting value of $h_j(t)$, which represents a measure of product quality, as $t$ approaches infinity, given that there is no R&D?

It approaches zero Signup and view all the answers

In the model, what is the relationship between the entry rate of non-generics, $n_j(t)$, and the parameter $\delta_j$?

The entry rate is not affected by $\delta_j$ Signup and view all the answers

What does the use of maximum likelihood standard error compared to the robust standard error indicate in the context of the given study?

The maximum likelihood standard error assumes a Poisson structure, while the robust standard error does not. Signup and view all the answers

How does using income and expenditure to compute market size affect estimates, compared to estimates based on a different methodology?

It tends to produce larger estimates compared to the estimates derived from alternative methods in the studied tables. Signup and view all the answers

What is the primary distinction between panels A and B in Table 3B, regarding estimation procedures?

Panel A uses maximum likelihood with no weights, while panel B uses a weighted maximum likelihood approach. Signup and view all the answers

What does the negative binomial model account for that the standard Poisson model does not?

It accounts for overdispersion of the Poisson parameter. Signup and view all the answers

What effect does using demographic information from multiple regions have on the study's market size measure?

It produces results that are similar to those using U.S. information alone, with no significant differences. Signup and view all the answers

If a potential market size measure based on the NAMCS data produces similar weighted and unweighted results, what does this suggest?

The distribution of the observations is more even across categories. Signup and view all the answers

In the context provided, what does the term 'overdispersion' refer to?

A situation where the variance of a dataset is larger than its mean. Signup and view all the answers

What is the main advantage of using a robust standard error over maximum likelihood standard error?

Robust standard errors are not limited by the assumptions of the chosen model. Signup and view all the answers

Study Notes

Market Size in Innovation: Theory and Evidence from the Pharmaceutical Industry

This paper examines the effect of market size on new drug entry and pharmaceutical innovation.
The study uses U.S. demographic trends to analyze exogenous market size changes.
A 1% increase in potential market size for a drug category correlated with a 4-6% increase in new drugs within that category.
The increase stems from both generic drug entry and new, non-generic drug entry.
The findings remain robust after controlling for various non-profit factors, pre-existing trends, and health care coverage changes.

Theory

A basic model links innovation rates to current and future market size.
Profit incentives and market size drive innovation, especially in the pharmaceutical industry.
Profit driven models and induced/directed technical change models consider profit incentives, market size, and specific innovations.
The study uses demographic trends as exogenous market size variations to overcome endogeneity issues inherent in other studies.
The analysis considers potential delays in drug development and approval processes.
This results suggest delays have no substantial effect but do affect calculations of the timing of R&D and patenting.

Empirical Strategy

The paper measures entry and innovation using FDA drug approvals.
A 1% increase in potential market size is correlated with an increase in new drug entry by 4-6%.
The robustness of the findings is reinforced by checking against lagged drug approvals, trends in biotechnology, insurance coverage, and economic incentives.
The study investigates whether current, past, or future market sizes have more impact on drug entry rates; current and short-term future market sizes yielded the strongest results.
The paper explores the relationship between market size and drug patents.
The analysis incorporates alternative measures of market size, including those based on demographic changes and health insurance coverage.

Generics and Non-Generics

The analysis extends to distinguish between generic and non-generic drug entry.
Introducing a new generic drug correlates with cost savings and improved availability, impacting market share and revenues.
Generic drug entry into an existing market is easier than introducing a fundamentally new drug, with lower required resources.
The study investigates whether different responses to market size changes exist between different types of drugs, and concludes that generics have a stronger response than non-generics.

Delays in Development

Delays in drug development and approval process are briefly addressed to mitigate potential biases.
The model accounts for delays using differential equations and shows that the response to market size is not significantly affected by delays.
Anticipation effects on drug entry are explored in a more general setting.

Supply-Side Determinants

The robustness of the findings is examined by controlling for non-profit factors, scientific incentives, and government funding (e.g., CRISP data).
No significant effects from any of the tested factors are observed on market size impacts.

Patents

The paper also examines the effect of market size on patents.
The analysis shows that the relationship between patents and market size is weaker and less statistically significant due to potential noise from various factors.
Evidence suggests that drug approvals are affected by past and anticipated market size, yet the results are less significant than other considerations.

Conclusion

The study finds a substantial and robust positive correlation between market size and new drug approval rates, both for generics and non-generics.
The response to market size changes is largely independent of factors like pre-existing trends in R&D, scientific funding sources, and health-related factors.
The results have implications for endogenous growth, directed innovations, and health related profit-based innovation incentives.

Studying That Suits You

Use AI to generate personalized quizzes and flashcards to suit your learning preferences.

Description

This study explores the relationship between market size and pharmaceutical innovation, particularly focusing on new drug entries. It reveals that larger potential markets significantly boost the number of new drugs, emphasizing the role of profit incentives. The analysis includes demographic trends to understand variations in market size affecting innovation rates.