Aggregate and disaggregate analysis of the effects of government expenditure on growth

Federica Calidoni

    Research output: Working paperDiscussion paper

    105 Downloads (Pure)


    This paper attempts to establish empirically whether different types of public spending are responsible for a decline in economic performance. Specifically, it investigates how total expenditure, and, in more details, government consumption and social security transfers, can affect productivity growth and, by consequence, the rate of growth of value added (VA). The first part of this work highlights the role of TFP in a country’s growth and its correlation with government expenditure. The regressions examine whether public spending contribute to a country’s growth through the marginal product of factors of production and their effects on TFP. Results vary considerably: with eleven manufacturing, total spending is significant and positively related to VA-growth, while the technological gap between countries appears insignificant. Adding to the analysis five services sectors, the outcome shows not significant correlation between government expenditure and growth performance. The second part includes a measure of openness calculated as the ratio of trade to GDP to test whether there is causal relationship between a country’s welfare and the size of its public sector or both public spending and growth depend on openness independently, and the direct statistical correlation between them is spurious; however there is no statistics of spurious correlation.
    Original languageEnglish
    PublisherUniversity of Dundee
    Publication statusPublished - 2004

    Publication series

    NameDundee Discussion Papers in Economics
    PublisherUniversity of Dundee
    ISSN (Print)1473-236X


    • Economic growth
    • Government expenditure
    • Total factor productivity

    Fingerprint Dive into the research topics of 'Aggregate and disaggregate analysis of the effects of government expenditure on growth'. Together they form a unique fingerprint.

    Cite this