Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit



Quantitative Methods: Econometrics A

Section A: Answer ALL questions

1. The impact of gender quotas in electoral lists

To address the scarcity of women in politics, in recent years many countries have      adopted gender quotas in candidate lists requiring the presence of a minimum share of female candidates. By construction, these quotas increase the share of female      candidates. However, it remains an empirical question whether they also lead to an increase in the share of women getting elected or reaching top political positions.     Let us focus in the case of local elections in Spain.  Within a proportional                      representation electoral system with closed lists, a quota requiring the presence of   at least 40% of candidates of each gender on the ballot was implemented in 2007 in municipalities with more than 5,000 inhabitants.

(a) One could use a difference-in-differences (DID) strategy to analyse the impact

of these quotas on the probability that a woman is elected as mayor. Explain  briefly how would you do it: write down the equation that you would                estimate for a simple DID and define all its terms. Explain also briefly what are the key requirements that would make the DID strategy adequate in this          context and potential threats to its validity. (3 marks)



Let us consider the possibility of using a regression discontinuity design (RDD) to estimate how quotas affect the probability that a woman becomes mayor. Explain briefly what are the key requirements that would make the RDD          strategy adequate in this particular context. (3 marks)

Let us now compare the RDD and the DID estimates. Do they identify the       impact of the treatment on the same type of municipalities?  Which one is      likely to provide more precise estimates? Which one is likely to provide more consistent estimates? (4 marks)

2. School closures have been among the most common non-pharmaceutical                  interventions to slow down the spread of the novel coronavirus.  Despite the large, expected costs of school closures for human capital formation, little is known about their impact. A recent paper titled “School re-openings after summer breaks in        Germany did not increase SARS-CoV-2 Cases” analyses how the opening of schools  affects COVID cases, exploiting the timing of school re-openings across German        states after the summer of 2020.1

In Germany, start and end dates of summer breaks differ between states to prevent  the entire German population from going on holiday at the same time.  The                 staggered summer breaks avoid traffic congestion as well as excess demand for          holiday accommodation in tourist regions.  The school year 2019/2020 ended as         planned between 22 June and 30 July followed by summer breaks of six weeks.  After the summer, schools were re-opened across all Germany, following this staggered      timing. The authors focus on this phase of full re-opening of schools in all states after the summer breaks which took place from early-August to mid-September 2020.        Figure 2 displays the spatial and temporal variation in school starting dates after the  summer break across German states and Figure 4 plots the point estimates and the 95% confidence intervals from a standard event study model comparing the                 evolution of the treatment and control groups.

Note: This graph shows a map of German counties highlighting the counties in states by date of school opening after summer vacation 2020. Counties (states) highlighted in dark grey start the new school year on the respective date,                                           while light grey indicates that they are still on summer vacation and medium grey      indicates that they had already re-opened schools at an earlier date. Schools               re-openings: 3 August: Mecklenburg-Vorpommern, 6 August: Hamburg,

10 August: Schleswig-Holstein, Berlin, Brandenburg, 17 August: Hessia,         Rhineland-Palatinate, and Saarland, 27 August: Lower Saxony, Bremen, and Saxony-Anhalt, 31 August: Saxony and Thuringa, 8 September: Bavaria,

14 September: Baden-Wuerttemberg. Source: KMK.

(a)         Would it be appropriate to use a difference-in-differences approach in order

to obtain a consistent estimate of the effect of school openings?  Discuss        briefly which assumptions need to be satisfied and whether you expect them to be satisfied in this context. (5 marks)

(b)         The authors include in their specification controls for changing mobility

patterns, as measured by Google Mobility Reports and commercial mobile phone data.  Please discuss under which conditions this is a valid control.

(5 marks)