Bank deposits and Google searches in a crisis economy: Bayesian non‐linear evidence for Greece (2009–2015)

Abstract

Due to a variety of reasons, the Greek economy faced a severe crisis being a member of EMU. Nonetheless, the country's banking system experienced a dramatic outflow of deposits, in the period 2009–2015. The present paper attempts to shed light on the possibility of forecasting bank deposits, based on the keyword “Grexit” of Google searches. In this context, apart from standard forecasting models like AR (p) and ARDL (p, q) we estimate a novel Neural Network ARDL (p, q, G) model and its respective Bayesian modification. We show that extending standard autoregressive models with the information provided by Internet Searches leads to significant improvement in the forecasting accuracy of the Bank Deposits, compared with other standard models. Furthermore, the forecasting performance of the models, which are extended with Google searches, is shown to be better than models containing only the well‐known indicators. Our findings are robust and econometrically sound.

Bank deposits and textual sentiment: When an European Central Bank president's speech is not just a speech

“…In a similar vein, Konstantakis et al. (2021) found that intensified Google searches containing the keyword “Grexit”, which summarize the public sentiment about the situation of the Greek economy, significantly affect depositors' perceptions and, ultimately their behavior regarding whether to withdraw their deposits or not. Lastly, Anastasiou and Drakos (2021), by introducing two modified Google search‐based crisis sentiment metrics of depositors' aggregate level of crisis sentiment, documented that depositors' crisis sentiment is related to irrational behavior and is negatively associated with bank deposit flows.…”

“…Many prior studies Drakos, 2020, 2021;Forssbaeck, 2011;Hasan et al, 2013;Konstantakis et al, 2021;Oliveira et al, 2011), have found that depositors exhibit low sensitivity to the bank or macro-specific fundamentals, with the effect being more prominent during crisis periods. Similarly, Hosono (2005) demonstrated that in South Korea, Malaysia, and Thailand, the sensitivity of deposit volumes to bank fundamentals declined after 1998.…”

“…One of them could be depositor sentiment. Although other types of sentiments have been explored in the deposits' literature (e.g., sentiment derived from Google searches; Anastasiou & Drakos, 2020; Anastasiou & Drakos, 2021; Konstantakis et al., 2021), we examine for first time the impact of the so‐called “textual” sentiment on depositors' behavior. This textual sentiment was derived from words steaming from the European Central Bank (ECB) President's speeches, and it is assumed to express information about the Euro area's economic and financial state (both current and future).…”

Consumers’ changing financial behavior during the COVID-19 lockdown: the case of Internet banking use in Greece

“…In this regard, a question arises: Why should we choose a country like Greece to apply this research? In this respect, in the last 10 years Greece has experienced a sovereign debt crisis, a bank-run and currently the pandemic (Featherstone 2011 ; Mink and De Haan 2013 ; Konstantakis et al 2021 ). The debt crisis led to a fiscal consolidation for almost 8 years, the banking crisis called for the imposition of capital controls in domestic and international capital flows, and the health crisis imposed a lockdown in almost all sectors in the Greek economy.…”

On the Construction of a Leading Indicator Based on News Headlines for Predicting Greek Deposit Outflows

“…They also found that these textual sentiments have a more significant impact on non-financial corporations than households. Konstantakis, Paraskeuopoulou, Michaelides, and Tsionas (2020) showed that extending standard autoregressive models with the keyword "Grexit" leads to improvement in the forecasting ability of the Bank Deposits in Greece, compared to models that contain just some standard explanatory variables. Their findings are checked for robustness in the data samples, in the activation function used and the prior distributions employed and are found to be econometrically robust and economically sound, confirming the crucial role of Google Searches for predicting bank deposits in the Greek economy.…”