O texto do Guilherme Macedo (hoje doutorando no Birkbeck College) e meu foi publicado no mais recente número da Revista de Economia Política.
This paper estimates the local multiplier of manufacturing for Brazil (2000-2010). The method is based on Moretti (2010) and on Moretti and Thulin (2012), who estimated these multipliers for the U.S. and Sweden. The local multiplier of manufacturing estimates the impacts of employment changes in the industrial sectors on employment in the servicessectors, and the impact of changes in employment in the high-tech and low-tech tradable sectors on employment in the services sectors. These estimates help to assess the importance of industrial employment changes over local economies. We created instrumental variables,based on the shift-share method. The employment data cover 21 economic subsectors and 123 regions in 2000, 2005 and 2010. We have estimated that in the Brazilian mesoregions, for each new job in the tradable sectors, almost four jobs were created locally in the services sectors. Additionally, each job in the high-tech industrial sectors was estimated to create approximately seven jobs in the services sectors over the long term.
Ainda na editoria de autopromoção deslavada: lembro que tiramos o 2o. lugar do Prêmio CNI com uma versão anterior desse texto.