The dynamics of public investment underpersistent electoral advantage.
This version: June 2011
This paper studies the eﬀects of asymmetries in re-election probabilities across par-ties on public policy and its subsequent propagation to the economy. The struggle be-tween opposing groups–that disagree on the composition of public consumption–resultsin governments being endogenously short-sighted: Systematic under investment in in-frastructure and overspending on public goods arise, as resources are more valuablewhen in power. Because the party enjoying an electoral advantage is relatively lessshort-sighted, it devotes a larger proportion of government revenues to productive pub-lic investment. Political turnover, together with asymmetric policy choices, induces
in an otherwise deterministic environment. I characterize thelong-run distribution of capital and show that output increases on average with politi-cal advantage, despite the fact that the size of the government expands as a percentageof GDP. Volatility, on the other hand, is non-monotonic in political power and is anadditional source of ineﬃciency.
JEL Classiﬁcation: E61, E62, H11, H29, H41, O23 Keywords:
Public Investment, Commitment, Probabilistic Voting, Markov Equilib-rium, Political Cycles, Time Consistency, Electoral Advantage, Ideological Bias.
An earlier version of this paper circulated under the title ‘On the dynamic ineﬃciency of governments.’I would like to thank Marco Battaglini, Steve Coate, Per Krusell, Antonio Merlo, and Pierre Yared foruseful comments. Part of this work was developed while visiting the IIES at Stockholm University and theUniversity of Pennsylvania. All errors are mine.
FRB of Philadelphia - Research Department. Email: firstname.lastname@example.org. The views ex-pressed in this paper are those of the author and do not necessarily reﬂect those of the Federal Re-serve Bank of Philadelphia or the Federal Reserve System. This paper is available free of charge atwww.philadelphiafed.org/research-and-data/publications/working-papers/