In my thesis I discuss and investigate whether there is an effect ofpoliticians' gender on policy outcomes. The main part consists of anempirical investigation of this question, using a rich set of paneldata on Norwegian municipalities.
In my empirical analysis I employ data on municipal budget sharesfor eight different service sectors and the share of women in thelocal council as well as various control variables. The data is fromthe period 1972-1999. I argue that it is particularly suitable forthe subject of study because female representation in politicsincreased rapidly in Norway during this period, partly due towomen's activism within the party system. This implies that thechanges in female representation it the local councils can be atleast partly considered as driven by forces outside of eachmunicipality.
Women in politics has been a popular subject of study in politicaleconomics during recent years, and investigating gender differencesin politics is of high interest in itself, both from a political andan academic point of view. However, my study is also related to acentral question in the field of political economics, namely whether(or to what extent) politicians are able to pursue their owninterests when deciding on policies.
Theoretical models differ in their predictions of the effects ofparty representation and politicians' personal ideology on policyoutcomes, and the evidence from empirical investigations is alsomixed. The traditional median voter theorem states that whoeventually gets elected does not matter at all for policies. Severalempirical studies lead to rejection of this 'full convergence'hypothesis. I offer a brief introduction to the theoreticaldiscussion of this subject and a somewhat more detailed review ofthe empirical literature, some of which focuses explicitly on femalepoliticians. I also review a few other studies which do not concernfemale politicians but gender differences in politics moregenerally.
A popular model showing that personal ideology could matter forpolitics is the citizen-candidate model, developed contemporaneouslyby Osborne and Slivinski and Besley and Case. Using a slightlyaltered version of this model, I show that if the barriers facingfemale candidates in the political system are lowered, this couldmake it more likely that a women is elected. This could again leadto different policy outcomes than when gender barriers are high,even though voter preferences have not changed. At the same time,the model shows that if voter preferences change towards a more'female' stand on policies, this could yield higher support forfemale candidates and ultimately affect the policy outcome. Themodel hence shows the importance of separating between these twoeffects.
My way of doing this in the panel data regression is to add a richset of electoral, demographical and socio-economic variables whichshould pick up effects related to voter preferences. In addition, Iadd municipality- and time- fixed effects (FE) to account forunobservable characteristics of each municipality and factors whichchange over time affecting the spending decisions in allmunicipalities.
The regressions are carried out using Stata 9. My results show apositive estimated effect of female representation in the localcouncil on the share of the budget devoted to childcare with a highdegree of statistical significance. This estimated effect does notdepend much on the choice of control variables, nor on the timeperiod of study. I also find a positive effect on the share spent onculture and a negative effect on administrative spending, thoughespecially the latter is less robust. The estimated effects on otherpurposes like for instance education and elderly care are eitherstatistically insignificant or highly unstable acrossspecifications.
The estimated effects are very small in magnitude. More precisely,the results imply that a ten percentage point increase in the numberof women in the local council only leads to an increase in thebudget share spent on childcare of 0.08 percentage points. Theeffect on cultural spending is of similar magnitude.
I argue that the robustness of the results to the inclusion of morecontrol variables could imply that there is not much of a problemwith omitted variables related to voter preferences. As arguedabove, changes in female council representation are likely to bedriven by external factors and could hence be considered exogenousin this setting. In an attempt to investigate this further, I tryestimating the model by two-stage least squares (2SLS) using theintroduction of gender quotation rules in the Norwegian Labour Partyas an instrument for the share of women in the municipality council.Instrumental variables using gender quotation rules in otherpolitical parties seem to be irrelevant.
Unfortunately, my instrument is to weak to give precise estimates ofthe effects on local public spending, and the 2SLS results thereforetell us nothing about the validity of the results from theregression without instrumental variables.
The results from the ordinary fixed-effects regression hence standuncontested: Increased female representation in Norwegian localpolitics seems to have had a positive impact on the composition oflocal public spending. The most robust effect is found for spendingon childcare, which is a sector where we might expect there to be anconflict of interest between women and men. This and the othereffects found are however very modest, possibly reflecting a largedegree of consensus rule in the Norwegian local democracy or heavyregulation of local public spending by the central government.