PHOTO CREDIT: websubs/Pixabay.com
Crises require fast responses by the state, no matter whether they follow from natural disasters, terrorist attacks, or the spread of diseases. Motivated by this rational, most constitutions equip their executives with extensive competencies to cope with emergency situations, so that governments can react on short notice and in a flexible manner. While the measures taken by governments to address the COVID-19 outbreak currently receive a lot of media attention, it is far less visible how parliaments respond to these special circumstances. However, legislatures have the right and obligation to hold governments accountable in ordinary and extraordinary times. In this blog contribution, I therefore explore how representatives have been overseeing the government in the COVID-19 crisis during the last weeks.
Building on principal-agent theory (Strøm 2000; Miller 2005), I expect that parliaments oversee government activities carefully during crisis situations. In all principal-agent relations, agents have informational advantages, making it difficult for principals to judge the appropriateness of decisions and actions taken on their behalf. If agents are forced to implement far-reaching measures under time pressure due to an emergency, the informational bias increases. Representatives, as principals of their government, hence should make use of instruments such as questions to members of the executive to understand what actions the government takes at what point in time and why during crises. In addition to information imbalances, agency slack characterizes all principal-agent relations, meaning that agents adopt measures that are not optimal for the principal. Again, one might expect that this problem is particularly pronounced in the COVID-19 crisis. Firstly, knowledge about the disease and, as a consequence, the evaluation of the appropriate instruments to fight it change quickly. Secondly, political actors need to consider a broad variety of consequences of the virus outbreak including individuals’ well-being, the functioning and spending of the health care system, but also effects on the economy, education, domestic violence and basic democratic freedoms. Representatives should be interested in ensuring that the executive adopts measures addressing all consequences of the virus that they deem important. And, given that there are often different strategies to approach a problem, they should try to make sure that the government’s action matches their preferred choice. Overall, parliaments should hence engage intensively into two aspects of legislative oversight: information retrieval and proposing to take (certain) actions to the government.
To test these expectations, I investigate the execution of legislative oversight mechanisms in six German states in the last three weeks. [1,2] In the German context, the state governments have the right to execute the most restrictive measures to cope with emergency situations. Some of the instruments set in place at the state level during the last weeks include the closing of child care facilities, schools, shops, bars, and restaurants as well as limitations of the freedom to assembly. In addition, state governments have far-reaching economic competencies that will allow them to provide financial support to businesses hit by the crisis. In consequence, the decisions reached by state executives are most influential for public life in Germany at the moment.
Figure 1 displays the case selection and their current affection by the COVID-19 crisis. The sample includes three states that have been strongly affected by the epidemic with rather high numbers of infected habitants already in early March (North Rhine-Westphalia, Baden-Wuerttemberg, Hamburg), one that currently displays a medium number of infections per habitants (Saarland) and two with still comparably low infection rates (Mecklenburg-Western Pomerania and Brandenburg). To date, the state assemblies continue their work despite of the infection rates, albeit under strict limitations. Some states cut the length of plenary debates and other face-to-face meetings to decrease the time MPs spend in one room, others reduced the number of participants or switched to digital communication for committee meetings. Spectators have been prohibited. However, only Saarland postponed its plenary meeting which was scheduled for early April, but still holds committee meetings. Mecklenburg-Western Pomerania, in turn, does not hold committee meetings at the moment. 
Figure 1 Total number of confirmed COVID-19 cases in German states reported by the Robert-Koch-Institute (for updates click the figure).
I study parliamentary questions (kleine Anfragen) as an indicator for information retrieval.  This instrument constitutes individual parliamentarians’ most powerful tool for police-patrol oversight (Saalfeld 2000) and reveals detailed information concerning legislators’ preferences and behavior (Martin 2011). Previous research already made use of the frequency of submitted questions as an indicator for the intensity of MPs’ monitoring activities (see e.g. Proksch and Slapin 2011; Jensen et al. 2013). I created a dataset including all questions submitted by MPs since March 2nd 2020 and published on parliamentary websites.
Overall, representatives submitted 178 questions to the government, but a mere 9 of them explicitly refer to the ongoing crisis (5.1%). I identify them by searching for the keywords COVID-19, Corona, infection, crisis, pandemic, and epidemic. The few questions engaging with the current situation distribute unevenly over the six states, ranging from a maximum of three in Hamburg and Baden-Wuerttemberg to a minimum of zero in Saarland and Brandenburg. Due to pronounced variation in the total number of questions submitted per state, the share of COVID-19 related questions amounts to a maximum of 23.1% (3 of 13) in Hamburg and 17.6% (3 of 17) in Baden-Wuerttemberg to 4.1% (2 of 49) in North-Rhine-Westphalia, 3.1% (1 of 32) in Mecklenburg-Western Pomerania and not a single one in Saarland (0 of 9) and Brandenburg (0 of 58).
Substantially, the small number of questions concerned with the crisis inquire about the state’s preparedness for the spread of the virus. They focus mostly on preparatory measures for the health care system (e.g. availability of safety clothing or the amount of intensive care beds in hospitals), but also include references to travel bans or the protection of citizens particularly vulnerable to the virus. Moreover, a few inquiries address measures to protect different branches of the economy from the consequences of the virus. Notably, not a single request in the time span under study explicitly engages with the measures concerning the closing of businesses, the ban of gatherings, or quarantine and isolation plans. Given that other countries have been fighting the virus through different strategies since January, it appears rather surprising that representatives did not submit any questions inquiring about governments plans to impose curfews or to cellphone tracking of infected in German states.
Even without explicitly referencing the ongoing crisis, a large proportion of questions might engage with topics that relate to the consequences of the current events for topics as diverse as the quality of education, childcare, or domestic violence. To reveal the substance of the questions submitted by MPs to the state governments, I calculate a structural topic model. Eleven topics emerged using a latent dirichlet allocation algorithm. The algorithm discovers sets of words that tend to appear together and clusters them into non-exclusive categories. The emerging topics for the requests submitted in March capture the main competences of the state level in Germany during day-to-day politics: Education, infrastructure, social protection and social care, environmental protection, public administration, and regional political issues such as large infrastructure projects. One of the categories engages explicitly with the COVID-19 crisis. There is hence no indication that topics implicitly related to the present event are gaining grounds in the parliamentary questions posed by MPs. Overall, this analysis suggests that, at least so far, parliamentarians in German states make only limited attempts to retrieve information from their governments that would allow them to judge the appropriateness and completeness of executive action during these difficult times.
Beyond information retrieval, MPs have the opportunity to force the government to take action on a certain problem through proposals. In a second data set, I gathered all of the proposals submitted to and published by parliamentary services since March 2nd 2020. During the three weeks, MPs submitted seventy-two proposals, of which 8.3% engaged with the current crisis. Again, the number in each state is rather low, albeit slightly more imbalanced than the distribution of the questions. I count three occurrences in Baden-Wuerttemberg (of 15), two in Hamburg (of 21), one in Brandenburg (of 3) and zero in North-Rhine-Westphalia (of 22), Mecklenburg-Western Pomerania (of 3), and Saarland (of 1). Notably, some of the states hit strongest by the virus seem to be more concerned with the topic as indicated by higher numbers of questions and proposals. However, this pattern is not consistent given the low oversight activities of MPs in North-Rhine-Westphalia when it comes to the crisis coping strategies of the government.
The substance of the proposals dealing with COVID-19 is similar to the requests. MPs ask the government to prepare the health care system for the crisis, to suspend compulsory schooling, or to revise the budget given the special circumstances. Two proposals submitted in Hamburg also demand the limitation of the duration of the latest emergency laws. In parallel to the questions posed by the MPs, neither of the initiatives addresses the measures taken by state governments in the last ten days, asks for additional or different crisis coping strategies such as far-reaching testing for the virus or a complete lockdown.
German state parliaments are not sleeping during the crisis. They try to continue with business as usual as far as possible without overseeing government actions concerning the emergency itself particularly closely. If representatives set the COVID-19 crisis on the agenda in the last weeks, they addressed a limited set of topics such as health care and how to compensate the economy. This insight is in stark contrast to my original expectation deduced from principal-agent theory. Principals do not oversee agents particularly closely in crisis situations. In fact, the representatives seem to cut the government some slack, leaving room for the executive to act without having to invest scarce resources into explaining decisions or reconsidering strategies.
By Corinna Kroeber in March 2020
 The analyses include the time between the 2nd of March (the day the Robert Koch Institute started to assess the health risk for the population as “medium”) and 23rd of March 2020 (the day this blog contribution was completed).
 Note that there were recently elections in Hamburg and that the first meeting of this parliament was only on March 18th 2020, meaning that opportunities to engage were limited for MPs.
 This information is based on the newsfeeds on state assemblies’ websites.
 Alternatively, one might look into question hours during plenary meeting, but only one of the six state parliaments held such an event to discuss the current crisis.
 Questions were forced into a total of 11 topics, which is the number of categories creating the most coherent topics for the data at hand when balancing maximization and minimization.
Jensen, C. B., Proksch, S.-O., and Slapin, J. B. (2013) ‘Parliamentary questions, oversight, and national opposition status in the European Parliament’, Legislative Studies Quarterly, 38 (2): 259-82.
Martin, S. (2011) ‘Parliamentary questions, the behaviour of legislators, and the function of legislatures: An introduction’, Journal of Legislative Studies, 17 (3): 259-70.
Miller, G. J. (2005) ‘The political evolution of principal-agent theory’, Annual Review of Political Science, 8 (1): 203-25.
Proksch, S.-O., and Slapin, J. B. (2011) ‘Parliamentary questions and oversight in the European Union’, European Journal of Political Research, 50 (1): 53-79.
Saalfeld, T. (2000) ‘Members of parliament and governments in western Europe: Agency relations and problems of oversight’, European Journal of Political Research, 37 (3): 353-76.
Strøm, K. (2000) ‘Delegation and accountability in parliamentary democracies’, European Journal of Political Research, 37 (3): 261-90.