BY MADDALENA CASTELLI
Maddalena Castelli is a MIEF student at SAIS studying International Economics and Finance. She plans to pursue a career in the private sector in New York.
Economists have long debated whether a more developed financial sector helps drive economic growth. King and Levine (1993) claimed “the predetermined component of financial development was a good indicator of long term growth.”Since then, changing circumstances have fueled arguments on both sides. To some, the 2008 global financial crisis suggested thatan abundance of financial intermediaries and insufficient regulations can lead to economic collapse. However, during the same time period, increased access to financial institutions in developing countries has had a profound impact on their growth—even though the majority of these transactions have occurred through non-traditional means like mobile phones.
The case for a linear relationship between financial development and economic growth originated with Joseph Schumpeter, in 1911. Schumpeter stated that “well-functioning banks spur technological innovation by identifying and funding entrepreneurs.” Furthermore, Schumpeter argued, a developed financial system will mobilize productive savings, allocate resources efficiently, improve risk management, and reduce information asymmetry, all of which facilitate innovation and entrepreneurship. Following Schumpeter’s theory,we would expect that as access to financial intermediaries increases, a country’s economic growth increases. Levine famously tested this relationship 20 years ago and found “a positive, significant, and partial correlation between the average annual rate of real per capita GDP growth and the average level of financial sector development” (King and Levine 1993).
In this paper, I revisit this theory and use more recent cross-country data to see if there is supporting evidence that financial development is a good indicator of long-term growth in today’s global economy.
To determine if the correlation exists, I examined cross-country data over 10 years, from 2005 to 2015. My analysis differs from Levine’s 1997 study in terms of time period, time span and the empirical model. From this new empirical framework, I was able to determine that while not as prominent as Levine had found in 1997, there is still a correlation between a country’s financial development and its rate of economic growth.
While the results from Levine’s 1997 study were robust, it has been 20 years since his research was originally published, and the global economy has undergone significant changes. The adoption of the Euro among European countries in 1999 removed the need for exchanging money and created a stable currency for the European Union. The North American Free Trade Agreement (NAFTA) in 1994 created a trilateral trade bloc between the United States, Canada and Mexico. In 2008, the world experienced the most severe financial crisis since the Great Depression. Lastly, we have witnessed the economic rise of China, which is currently the second largest economy in the world. These events represent a trend toward greater global economic integration and could affect the relationship between financial development and growth.
To revisit the relationship Levine put forward, I first gathered the per capita real GDP growth of 87 countries. Secondly, to determine if the level of financial development played a role in real GDP growth, I looked at three different financial indicators: access to financial intermediaries, depth of credit, and loan stability. Thirdly, I used openness to trade, school enrollment, government consumption, and initial GDP per capita as control variables.
My cross-country data is categorized into four income groups: low (23 countries), lower-middle (26), upper-middle (21), and high income (17), in which I examine the per capita real GDP growth (constant LCU). These 87 countries were chosen due to the availability of data for both the financial development and economic growth indicators.
To understand the level of financial development for each country I looked at three indicators:
- Access: Banks per 1000 adults. Typically, higher levels of access to and use of banking services can expand opportunities.
- Depth: The ratio of private credit by deposit money banks to GDP (%). Depth measures the size of financial intermediaries.
- Stability: Bank nonperforming loans to gross loans (%). Understanding how many defaulting loans a bank is holding allows me to explain the role that Depth and Access have on the overall health of the financial intermediaries of the country
Instead of just looking at how the three financial development indicators affect per capita growth, I added four control variables:
- Trade: I measured openness to trade using imports plus exports over real GDP. More open countries are expected to grow faster for a number of reasons, including comparative advantage, technology and knowledge transfers, and the adoption of improved regulatory standards through trade agreements.
- School Enrollment: To measure the impact of human capital, I controlled for the level of secondary school enrollment.
- Government Consumption: To understand the role that the government plays in each country, I looked at the final government consumption expenditure.
- GDP Constant: Initial GDP per capita could affect growth through conditional 𝛽-convergence. The concept of 𝛽-convergence is based on the idea that growth rates are inversely related to countries’ initial level of production per worker—essentially that poorer countries grow faster, allowing them to catch up to, or “converge,” with wealthier countries.
To examine the strength of the empirical relationship between financial indicators and growth averaged over the 2005-2015 period, I constructed the following OLS regression:
Where G is the value of growth indicator (per capita GDP growth), F(i) is the value if the ith financial development indicators (Access, Depth, and Stability), and X(j) is the value of the jth control factors associated with economic growth (Trade, School Enrollment, Government Consumption, and GDP Constant).
I would expect a positive relationship between Growth, Access, and Depth, and a negative correlation between Growth and Stability. As access to financial intermediaries increases, a country’s economic growth is expected to increase, which is similar to the hypothesis articulated by Schumpeter and tested by Levine.
Even though my results are not as robust as Levine’s, the variables Access, Depth, Trade, and School Enrollment are still statistically significant for different income groups. A possible explanation for Access being significant for the lower-middle income group is the role of inclusion in growth. Limited access to capital represents a significant challenge for economic growth and stability. Gaining access to financial services is critical to support development in low-income countries. Depth is statistically significant at the 5% level for the upper-middle income group. Financial deepening is likely to raise the ratio of money supply to GDP, making more funding available for growth and investment. Upper-middle income countries appear to benefit the most from increased financial openness in order to stimulate growth and expand markets. As expected, when Stability, which measures the proportion of nonperforming loans,increases, economic growth decreases. (I eliminated this variable when running the regression model for low-income countries because there were not enough observations.)
Among the control variables, Trade is statically significant at 5% for low income countries. Although trade openness in developing countries is associated with higher inequality, it may mean only that higher incomes are rising faster than lower incomes – not that lower incomes are falling. Over time, trade and technological transfers may push the economy to become more productive and change its structure, resulting in lower inequality. School Enrollment is positive across the four income groups, although only significant for the low-income group, reinforcing the positive correlation between years of schooling and faster growth.
The table above shows evidence of conditional 𝛽-convergence (𝛽 <1) in low and upper-middle income groups. Typically, initially richer countries grow more slowly than initially poorer countries after controlling for the initial real GDP per capita. In general, the data presented in this research supports the initial hypothesis that increases in financial development is a good indicator of economic growth.
While economists continue to investigate which factors drive growth, there is no universal consensus on which are the most important. However, this research indicates that financial development plays an important role. As originally articulated by Schumpeter, a well-functioning and developed financial system spurs technological innovation.The availability of finance allows resources to be directed towards productive activities that might not have been undertaken if there had not been the option to invest. In less developed financial systems, frictions such as information asymmetry limit the amount of credit available for the expansion of newly productive firms, lowering productivity and output per worker.
Even though the economic landscape has changed significantly since Levine (1997) concluded that his analysis “demonstrates a strong positive link between the financial system and long term economic growth,” I arrived at the conclusion that there is positive evidence of the relationship, although it is not as pronounced as the previous literature states. I was also able to conclude that depending on the income group, Access, Depth, and Trade are statistically significant at 5%. There is also evidence of convergence; poor countries tend to grow faster than rich ones, implying that the income gap is narrowing.
My analysis also found that the values of Access and Depth have increased in most countries among the four income groups. The underlying reason might be the growing emphasis on inclusive growth by policymakers around the world. For instance, Myanmar’s financial sector authorities have made steps toward a regulatory environment that allows microfinance institutions and mobile network operators to provide international and local remittances.The analysis suggests that these changes are associated with greater GDP growth.
While my analysis shows a positive relationship between financial development and economic growth, more research needs to be conducted in order to determine causality between the two. Although there is a two-way relationship, it is possible that faster-growing countries tend to place more emphasis on financial development. We also need to better understand the role of other variables such as macroeconomic policies and regulations, legal, and geographic characteristics in determining economic development. As Levine observed in his conclusion twenty years ago, the empirical evidence does not link specific financial sector policies with long term growth.
Barro, Robert J. (1991). “Economic Growth in a Cross Section of Countries.” Quarterly Journal of Economics 106 (2): 407-443, https://doi.org/10.2307/2937943
Cihák, Martin, Asli Demirgüç-Kunt, Erik Feyen, and Ross Levine. (2013). “Measuring financial development.” VoxEu. http://voxeu.org/article/measuring-financial-development
Domestic credit provided by financial sector (% of GDP) by Country. (n.d.). https://www.indexmundi.com/facts/indicators/FS.AST.DOMS.GD.ZS Sources: International Monetary Fund, International Financial Statistics and data files, and World Bank and OECD GDP estimates
Greenwood, Jeremy, and Boyan Jovanovic. (1990). “Financial Development, Growth, and the Distribution of Income.” Journal of Political Economy 98 (5): 1076-1107http://piketty.pse.ens.fr/files/GreenwoodJovanovicJPE1990.pdf
Hanushek, Eric, and Ludger Wößmann. (2007). Education Quality and Economic Growth. Washington, D.C.: World Bank Group http://siteresources.worldbank.org/EDUCATION/Resources/278200-1099079877269/547664-1099079934475/Edu_Quality_Economic_Growth.pdf
King, Robert, and Ross Levine. (1993). “Finance and Growth: Schumpeter Might be Right.” Quarterly Journal of Economics 108 (3): 717-737
Levine, Ross. (1997). “Financial Development and Economic Growth: Views and Agenda.”Journal of Economic Literature 35: 688-726.
Loayza, Norman, and Romain Rancière. (2004). “Financial Development, Financial Fragility, and Growth.” https://openknowledge.worldbank.org/bitstream/handle/10986/14233/wps3431.pdf?sequence=1&isAllowed=y
Rioja, Felix, and Neven Valev. (2003). “Does one size fit all? A reexamination of the finance and growth relationship.” Journal of Development Economics 74: 429-447. http://www2.gsu.edu/~ecofkr/papers/pub2004.pdf
Samargandi, Nahla, Jan Fidrmuc, and Sugata Ghosh. (2014). “Is the Relationship between Financial Development and Economic Growth Monotonic? Evidence from a Sample of Middle Income Countries.” CESifo Working Paper No. 4743. http://www.cesifo-group.de/DocDL/cesifo1_wp4743.pdf