Score-driven latent-factor panel data models of economic freedom: an empirical application to the United States

ABSTRACT

In this paper, we study the link between economic freedom and gross domestic product (GDP) growth of 12 industries making up the United States (US) economy for 50 US states from 2005 to 2020. To measure the industry-specific impact of economic freedom in the US, we use a novel panel data model, named the score-driven latent-factor panel data model of economic freedom, which includes US state- and industry-specific score-driven components, US state- and industry-specific unobserved effects, and federal-level latent factor. We show that the statistical performance of the novel panel data model is superior to those of classical static and dynamic panel data models. With the exception of the ‘Agriculture’ and ‘Utilities’ industries, we find a positive relationship between economic freedom and growth in 10 of the 12 US industries considered for the score-driven latent-factor panel data model.

KEYWORDS:

• Economic freedom
• unobserved effects panel data models
• dynamic conditional score (DCS) models
• generalized autoregressive score (GAS) models
• efficient importance sampling (EIS)

JEL CLASSIFICATION:

• C33
• O47

Acknowledgements

We appreciate the helpful comments and suggestions of the editor and reviewers of the journal. This paper was presented at the 23rd Annual Academic Conference of the Society of Business, Industry, and Economics, Destin, Florida, April 2023. We greatly appreciate the comments of the conference participants. Data and computer codes are available from the authors upon request.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/00036846.2024.2354515

Additional information

Funding

We appreciate funding from the Center for the Study of Economics and Liberty at Mercer University.