Bureaucratic reputation is one of the most important concepts used to understand the behaviour of administrative agencies and their interactions with multiple audiences. Despite a rich theoretical literature discussing reputation, we do not have a comparable measure across agencies, between countries, and over time. I present a new strategy to measure bureaucratic reputation from legislative speeches with word-embedding techniques. I introduce an original dataset on the reputation of 432 bureaucratic bodies over a period of forty years, and across two countries, the US and the UK. I perform several validation tests and present an application of this method to investigate whether partisanship and agency politicisation matter for reputation. I find that agencies enjoy a better reputation among the members of the party in government, with partisan differences less pronounced for independent bodies. I finally discuss how this measurement strategy can contribute to classic and new questions about political-administrative interactions.
Models of bureaucratic accountability rest on the ability of politicians to obtain information about bureaucracies. Yet how politicians process this information is an understudied and central topic. I argue that politicians follow partisan motives when updating their beliefs about bureaucracy. I leverage a new measurement strategy and apply word-embedding techniques to a corpus of legislative speeches and estimate partisan beliefs about 323 agencies over 40 years in the UK and the US. l build panel data models and a generalised difference-in-differences design and find that, when there is party-executive alignment, beliefs are on average more positive by 3-4 percentage points, irrespective of the party-agency ideological distance and the agency's level of politicisation. I show how bias follows from beliefs to behaviour. When co-partisan with the executive, (1) politicians' speeches about agencies are less grounded in facts and evidence, (2) and politicians are less likely to oversee ideologically distant agencies.
We study the consequences of populism on bureaucratic expertise and government performance. We use novel data on about 8,000 municipalities in Italy, over a period of 20 years, and we estimate the effect of electing a populist mayor with a close-election regression discontinuity design. We find that the election of a populist mayor leads to (1) higher turnover among top bureaucrats; (2) an increase in the probability of replacing expert with non-expert bureaucrats; (3) a decrease in the percentage of highly educated bureaucrats; (4) and lower performance overall. Moreover, we find evidence that the increased inefficiency of the bureaucracy is accompanied by proliferation of council and executive resolutions, in line with the recent literature on overproduction of laws and bureaucratic inefficiency.
The political control of the bureaucracy remains a classical topic in political science. However, little is known about its reverse: bureaucracies influencing politicians. I conceptualise bureaucratic influence as the extent to which legislators use the information produced by agencies in the legislative process. Building on cheap talk models of strategic communication, I argue that legislators make greater use of bureaucratic information when ideologically closer to agencies and that agency independence – operating as a credibility-enhancing mechanism – mitigates the effect of ideological distance. I introduce a new measurement strategy to estimate legislators' use of bureaucratic information which employs syntactic analysis and apply it to a corpus of 6.8 million speeches given by US congresspersons in floor and committee speeches. I find that ideological distance reduces politicians' use of bureaucratic information, but only in floor speeches. This suggests that in congressional committees, and less partisan arenas, ideology matters less for bureaucratic influence.
The literature on government composition and fiscal policy is unanimous on theory though discordant on findings. Known as the fiscal commons problem, theory predicts more politicians lead to inefficiencies, but findings are mixed. With new data on Italian municipalities, I estimate the effect of government composition on a battery of planned and actual budget outcomes: expenditures, revenues, taxes, financial transfers, and deficit. I leverage two reforms implemented in 2011 and estimate treatment effects with a difference-in-differences and a difference-in-discontinuities design. I find that more politicians lead to increased planned spending and deficit, which are nonetheless coupled with larger revenues. However, the effects disappear when looking at actual budget items. I find that more politicians actually lead to less deficit and have no effects on spending. I present supporting evidence in favour of an electoral incentive-based mechanism and rule out alternative explanations centred on party competition.