On recessive and expansionary impact of financial development: empirical evidence.

• Author: Nguena, Christian-Lambert

1. Introduction

   How can we explain the cyclical fluctuations of the economy in general and the economic recession in particular? This is one of the most addressed and controversial issues among economists. According to Minsky (1975), the cumulative chain of increasing oversupply and insolvency of borrowers was probably at the heart of the "Great Depression" of the 1930s; this cumulative chain is also considered as a main factor explaining the "Great Recession" of the 2000s. This assertion of Minsky (1975) is based on Fisher's (1933) analysis of "debt deflation". According to him, the decrease in prices due to free-market play can trigger a deflationary spiral, leading to insolvency and chain bankruptcies; indeed, deflation increases the real value of corporate debt. To reimburse, they stop investing and hiring and destock greatly. This rational makes it important to focus on finance and its impact on the recession through its role in triggering crises. Since the invention of finance, it has been associated with episodes of excessive risk-taking and financial asset price bubbles (Nguena & Tsafack Nanfosso, 2014). Such episodes usually had a bad end, causing job and income losses far beyond the financial industry. Considering this, it is justified to verify the assumption of whether solving the problem of borrower insolvency through private domestic credit development would help or not combat the recession.

   A recession is different from a simple slowdown in the economy that corresponds to a decline in the GDP (1) growth rate over two consecutive quarters. The National Bureau of Economic Research, through the Committee to date business cycles define it as a "significant decline in economic activity affecting all sectors, for several months, normally visible in production, employment, real income and other indicators. A recession begins when the economy reaches a climax and ends when it reaches its lowest point". The term depression, on the other hand, refers to a more intense and long-lasting fall in production as experienced in the United States after the stock market crash of 1929.

   Financial development for its parts involves improvements in the following functions provided by the financial systems: (i) pooling of savings; (ii) allocating capital for investments; (iii) monitoring investments; (iv) risk diversification; and (v) exchange of goods and services (Levine, 2005). Moreover, funds allocation, savings and investment decisions are sensitive to each of these financial functions. As a consequence, total factor productivity along with physical and human capital accumulation are the main channels through which finance impact economic growth (Tsafack Nanfosso & Nguena, 2015; Nguena et al., 2021). With the possibility of reduction of information asymmetries and insurance of risk sharing, the financial sector can absorb shocks and reduce the amplification of cycles via the financial accelerator, and thus lowering inequality volatility (Bernanke et al., 1999).

   While it is true that developing countries have experienced very few depressions throughout the years, these countries faced and are facing many types of crises; the current pandemic crises is the last ever-lived phenomenon with effective and potential recessive and/or expansionary impacts. Before it, the global economy tipped into recession in 2008 due to the 2007 financial crises (2). This recession has driven firstly a contraction in demand from households and businesses due to the tightening of credit standards by banks and the fall of their earned incomes, resulting from the rise in unemployment and the collapse of world trade (Redoules, 2009). In such context, policymakers need to set robust diagnostic to comprehensively assess the impact of recent and past crises on the real economy, and the economic impact of financial development that would insidiously be associated with an underlying recession.

   This brief theoretical and empirical presentation about the necessity to combat the recession and the importance of financial development implies the necessity to check the existence and the nature of the relationship between both phenomena. Thus, to contribute to the existing literature, the issues that this paper addresses is to qualitatively and quantitatively question the impact of financial development on economic recession. Practically, we examined this question using data for 129 countries covering the period 1990-2010 along with a plural and innovative methodology step by step.

   We developed a Sasabuchi test to verify the inverse U-shape and estimate the extreme point for the non-linear specification. For robustness check, we used (1) an Iteratively Reweighted Least Squares method to control the extreme values that could influence the baseline findings and (2) an exclusion of regions technic to control the heterogeneity of the sample and assess the behaviour of the variables of interest. Indeed, given the heterogeneity of our sample, we considered a global sample of developing countries and a grouping sample based on regional ralations; this is mainly because we assume that financial development is correlated with the level of wealth and present divergent characteristics across regions (Demirguy-Kunt & Levine, 1999; Beck et al., 2014, 2016; Nguena, 2020).

   Moreover, we implemented also "Feasible Generalized Least Squares", "Locally Weighted Scatterplot Smoothing" and "Local Linear" regression methods. Several estimations based on different specification using a semi-parametric regression with a procedure provided by Lokshin (2006) were executed. The estimation output consisted of two parts: (1) a table that reports the regression coefficients of the linear part of the model and (2) a graph that illustrates the functional form of the nonlinear part of the relationship between both indicators.

   Such an investigation has numerous interests. Firstly, the focus on seeking solutions on hold issue such as recession linked to the past economic crisis and depression (The "Great Depression" of the 1930s, the "Great Recession" of 2007-2009 followed by the last financial crisis) and, more interestingly, to the current health crisis which is starting to imply an economic crisis. Secondly, we present a literature review and especially a theoretical framework on the important question of financial development and its link with the economic recession. Thirdly, our empirical methodology is based on a sample of 129 developing countries from all regions of the world.

   Fourthly, we use primary and hitherto, almost unexploited "Rare macroeconomic disasters" data from Barro & Ursua (2012); this database permits us to have a specific proxy of "economic recession" closer to the reality; compared with data from other sources (IMF (3), World Bank) this recent dataset on economic recession from the said sources should provide findings with updated and more focused policy implications. Fifthly, as far as we have reviewed, no study has employed the empirical approach we use in tackling the problem; thus, the paper also contributes to the existing literature on the methodological aspect. Finally, given the challenges of globalization, especially considering financial sector performance, the analysis could serve as a basis for the implementation and/or revision of policies towards banking and financial system development by authorities/international organisations.

   We mainly found strong evidence of the nonlinearity and U-shaped relationship along with a threshold effect. The semiparametric regression shows that the results of the parametric part converge with the previous results in general and shows with illustration the functional form of the nonlinear relation between recession and financial development. Verifying the robustness permitted us to confirm the results of the baseline and extended model specification in terms of coefficients sign and significance.

   Controlling the heterogenity within the sample permitted to highlight SSA, SASIA and LAC as the order of continental/regional importance in increasing magnitude, demonstrating that fuels for South Asia (SASIA) and Latin America and Caribbean (LAC) countries and financial openness for sub-Saharan Africa (SSA) countries are negatively related to recessions. The control of extreme values confirmed the baseline findings and highlighted the following as the order of regional importance in increasing magnitude: sub-Saharan Africa (SSA), South Asia (SASIA) and Latin America and the Caribbean (LAC). Moreover, the turning point permitted to divide the sample into two parts: on one hand, countries or regions which are eligible to an expansionary impact of financial development, and on the other hand, countries or regions with a recessive impact of financial development.

   The rest of the paper is organized as follows: In section 1, we present the updated literature review on the subject; in section 2, the theoretical framework and stylized facts; in section 3 we specify the model, discuss the data as well as sensitivity analyses for robustness checks. The empirical analysis is presented in Section 4. Section 5 concludes with policy recommendations.

2. Recessive and expansionary impact of finance: a literature review

   Firstly, we start by the presentation of an overall up to date empirical studies analysing the relationship between financial development and economic recession and; secondly fall into a classification, step by step based on major results in the literature and different methodologies used and/or geographical orientation.

   2.1. Empirical studies on the relationship between finance and economic recession

   When a country is in recession, all the sectors affected are normally visible in production, employment, real income and other indicators. This is the main reason why, the literature on the relationship between financial development and economic recession, usually analyses the effect on the previously stated...