Central government behaviour is modelled as a function of variables reflecting benevolent welfare maximiser intentions as well as those reflecting re-election motives. For checking what is affecting the chances of grant receivals (of any applicant from a municipality or of local government itself) several Probit and linear models have been tested with different sets of political and socio-economic controls on a combined dataset for all Hungarian municipalities (n=3168). This period (starting with the country’s 2004 EU Accession) spans three election cycles (2002-2006; 2006-2010, 2010-2014). Estimations are carried out on the whole and sub-samples by size and different pre- and post-election periods. Results show partisanship elements (same political colour favouritism), and reinforce EU SF literature: efficient usage of EU funds depends mostly on institutional conditions- since here proxies for local administrative capacity and earlier EU project experience are strongly significant and positive. Socio-economic and need controls show a mixed picture, reflecting the conflict of efficiency vs. equity-driven policy goals of development policy today.