European banks are not immune to national elections

Do elections affect the lending behavior of European banks? Do European banks change credit supply and loan pricing as major national elections approach? Employing a unique monthly dataset that draws on the data covering some 250 banks in 19 Eurozone countries from 2010 to 2020, we find that European banks increase their loan rates for corporate and housing loans ahead of elections. This finding supports the view that loan pricing changes of European banks reflect the electoral uncertainty inherent in the democratic election process. The impact of elections is more pronounced for small banks. Furthermore, we obtain some evidence that elections affect the credit supply of banks, especially when elections are preceded by high uncertainty. Our results provide useful insights about changes in the lending behavior of banks related to the electoral calendar.

Bank lending and elections

With over half of the world’s population scheduled to vote in an election this year, more people than ever could vote.1 Elections are naturally important in deciding about the composition of new governments and forthcoming policies, but election periods may also influence the behavior of various market participants. In this note, we focus on bank lending behavior in the months before elections.

Two views dominate the discussion on how banks adjust their lending behavior with respect to the credit supply and loan pricing in the run-up to major national elections. The manipulation view assumes that bank lending increases before an election. This view aligns with the political economy literature, whereby politicians try to manipulate economic instruments to enhance their chances of re-election. A handful of empirical studies support this view by showing increases in state-owned bank lending in election years in emerging countries (e.g. Dinc, 2005; Carvalho, 2014; Bircan and Saka, 2021). Such evidence has also been identified for developed countries such as Germany (Englmaier and Stowasser, 2017) and France (Delatte, Matray, and Pinardon-Touati, 2020). On the other hand, the bulk of evidence on emerging countries and state-owned banks suggests this view to be generally less applicable to European countries. Compared to emerging countries, banks in Europe are more likely to be privately owned and state institutions more robust. Such settings should reduce the ability of politicians to alter the pre-electoral behavior of banks and thus the relevance of the manipulation view.

Under the alternative uncertainty view, banks face heightened uncertainty, making them unsure about the election outcome and future economic policies. In response, bank lending decreases and loan prices increase as the election approaches. Several studies confirm higher loan pricing (e.g. Francis, Hasan and Zhu, 2014; Ashraf and Shen, 2019), as well as reduction in the supply of mortgage loans in the case of the US (Kara and Yook, 2023). Thus, the uncertainty view seems more appropriate for democracies, which are characterized by less predictable electoral outcomes.

In our recent paper (Fungáčová, Kerola and Weill, 2024), we study whether major national elections affect the supply and the interest rate of new loans of European banks. As the majority of studies investigating political interference in lending behavior of banks before elections rely on data from emerging countries, our investigation contributes to understanding the behavior of banks in developed countries.

Loans become more expensive before elections

Employing a monthly dataset for roughly 250 banks from 19 Eurozone countries that covers the period from January 2010 until December 2020, we study bank lending behavior at specific points in the run-up to elections. Unlike earlier studies, our dataset allows us to distinguish among corporate, housing, and consumption loans. The dataset consists of Eurozone countries, so all banks in our sample operate under the same monetary policy which can thus be considered exogenous to the forthcoming elections in our framework due to the ECB’s political independence. We consider the main national elections in each country.

We first investigate whether elections affect the loan rates for different types of loans. The results indicate that European banks tend to increase their prices on corporate and housing loans in the months before elections. This finding comports with the uncertainty view; loan pricing changes of European banks are driven by the electoral uncertainty inherent to the democratic election process. Moreover, the positive effect of upcoming elections on loan pricing mainly comes from small banks, the group most affected by the uncertainty surrounding elections.

We also consider a possible relation between elections and the amounts of new loans. In this case, we find significant results only for housing loans. The amount of new housing loans decreases in one to three quarters ahead of elections, so only the credit supply of this type of loans is affected by the occurrence of elections in European countries. This result is most pronounced for smaller banks. Once again, these findings favor the uncertainty view.

How much uncertainty matters?

The level of uncertainty varies considerably before an election depending on the country and the time to election. A heightened level of uncertainty before elections could spur higher lending rates and decrease the volume of new loans issued. To investigate this question, we use Google Trends data and create a measure for political uncertainty around elections. Our measure is based on the assumption that people are likely to seek information online when they are uncertain. Thus, an increased number of web searches would indicate higher uncertainty. Based on our Google Trends measure, we identify periods of high political uncertainty and examine whether the lending behavior of banks before election changes. We find that the positive effect of elections on loan pricing and the negative impact on new loan volumes are strongest in the presence of high uncertainty. These results are in line with our expectations and provide support for the hypothesis that the level of uncertainty from an upcoming election affects the lending behavior of banks. This evidence also bolsters the uncertainty view.

Our findings provide useful insights concerning changes in the lending behavior of banks linked to the electoral calendar. The influence of elections on loan rates and loan amounts of certain types of loans suggests that elections can influence the effectiveness of monetary policy and thereby contribute to heterogeneity across Eurozone countries.