Mark Partridge (p), SH Chung, Oudem Hean

Marcos Sanso-Navarro

In the late 1990s, California led the way in legalizing medical marijuana, beginning a trickle of states allowing medical sales. This was followed by Colorado and Washington legalizing recreational sales marijuana in 2013/2014. Since then, there has been an avalanche of states legalizing both medical and recreational marijuana, with nearly 20 states legalizing recreational sales by early 2023. Until now, it has been difficult to assess the medium- to long-term economic effects of recreational marijuana legalization because the length of time since enactment was too short. This study takes advantage of longer elapsed time to assess the economic impacts in several states that legalized recreational sales by 2018. The empirical methodology employs fixed-effects regression using county-level data and synthetic control matching for individual states (SCM).

The results indicate that there are weak positive effects of legalizing recreational sales on overall state employment, wages, housing costs, and the share of the population that is young adults. Using the spatial equilibrium model, these findings further suggest that both productivity and quality of life—especially for young adults—improved. The results also indicate that early adopters benefited more than those that followed. Furthermore, the regression results suggest that within states that legalized, counties that “opted-in” and allowed recreational sales benefited more than counties that did not allow recreational sales, further supporting the notion that legalization has weak positive economic effects through enhancing quality of life. Those who “opted-out” of sales were unable to capture tangible positive spillovers by “freeriding” on their local neighbors that approved sales.

Judit Berkes (p), Tamás Dusek

Mark Partridge

The aim of the study is to analyse the changes in the economic competitiveness and performance of Hungarian regions in the last 20 years. In the first part of the paper, the concept of competitiveness, and in particular the concept of regional competitiveness, will be critically examine and summarise, as well as the “testing spatial convergence” approach to the temporal analysis of spatial economic development. The empirical analysis will be conducted at the county level (NUTS 3 level, 20 spatial units), including a wide range of economic, social and demographic variables, and examining the impact of regional development programmes and large infrastructure investments, also attempting to explain the varying degrees of regional change. Our methodology combines the analysis of individual indicators with the use of complex aggregated indicators composed of several indicators.
Some of the counties – such as the central and north-western part of the country – show higher level of competitiveness with well-developed infrastructure, high quality education and a strong economic base. These counties also offer greater opportunities for investors and workers and internal migration also shows their greater attractiveness. Other counties, such as some in the eastern part of the country, are less developed and may be less attractive to investors and workers. However, the competitiveness picture is not entirely stable, especially at the level of larger regions of the country, and the position of neighbouring counties has changed significantly in some cases.

Marcos Sanso-Navarro (p), Jose Olmo, María Vera Cabello

Judit Berkes

This paper extends the definition of time series convergence beyond the conventional parity condition by allowing for positive cointegration between pairs of unit roots. The paper also develops a novel one-sided test of time series convergence and presents its asymptotic properties under the null hypothesis of no convergence and the alternative hypothesis. The test is robust to general forms of weak dependence in the transitory components and does not require the estimation of the cointegration coefficient. These features are illustrated in a Monte-Carlo simulation exercise for a battery of ARMA(1,1) innovations of the unit root processes. As a byproduct, we propose a methodology to detect convergence clubs. This procedure is based on centrality measures of network dependence given by the degree and betweenness. The empirical application analyzes regional data on population and per-capita income at the NUTS-2 level from France, Italy and Spain. Our results uncover the presence of different convergence clubs for population dynamics and convergence to a single regime for per-capita income.