Financial Development, Financial Openness, and Policy Effectiveness

Abstract

This study explores how financial development and openness influence the effectiveness of fiscal and monetary policies. An analysis of data from about 100 countries between 1980 and 2018 reveals that both financial openness and development weaken the impact of monetary and fiscal policies. Our results further show that financial development in a country diminishes policy effectiveness depending on the country’s level of financial development; specifically, the more developed a country, the less effective the policies would be. Additionally, through a detailed examination employing a dynamic panel GMM approach, the study investigates the global repercussions of economic downturns in the US and how financial maturity shapes policy effectiveness during these times. We also discuss some policy implications that show that the positive impacts of monetary policy on output growth are lessened during crisis periods, and policymakers should act accordingly.

1. Introduction

The global financial system has significantly advanced over the last few decades, partly due to increased international trade and globalization. Such advancements have increased access to credit through commercial banks, mutual funds, co-operatives, etc., thereby increasing the money supply in economies worldwide. Additionally, governments worldwide have actively been involved in various bilateral and multilateral trade and financial agreements to facilitate the cross-border flow of goods, services, and capital. These advancements have changed the way policies and business activities are undertaken. The literature suggests that financial development and openness have enormous economic benefits, as these variables are associated with economic growth (see, for example, Levine 2005; Hassan et al. 2011; Estrada et al. 2015). Despite such benefits, some studies suggest that a country’s financial development level has some adverse effects on the economy, which cannot be discounted. For example, Kose et al. (2008) show that many nations experience increased volatility and more severe external shocks due to increasing financial openness, underscoring the importance of strong financial institutions. Panizza (2013) studies the association between economic growth and financial development across countries and provides evidence that a higher degree of financial development often leads to increased inequality and heightened exposure to financial crises. In this context, revisiting the association between a nation’s financial development, financial openness, and economic growth is a worthy undertaking.

It is well documented in the literature that a well-functioning financial system is crucial to conducting monetary and fiscal policy. The study of the association between economic development and the financial system dates back to Gurley and Shaw (1955), followed by Taylor (1987), Mullineux (1994), and recently, Ma and Lin (2016). Of late, however, following the work by Bernanke and Gertler (1995), an increasing number of papers have relied on the credit channel theory of business cycle fluctuations to explore the association between a nation’s financial institutions and economic development. The recent papers that focus on this channel include Nam et al. (2021), Gertler et al. (2020), and Abbate et al. (2023). The results from these studies largely conclude that while financial frictions amplify the effects of adverse exogenous shocks on the economy, they dampen the impact of policy shocks.

Like financial development, financial openness is another crucial variable that represents the strength of financial institutions in a country. Many bilateral and multilateral trade agreements and global institutions like the International Monetary Fund and the World Bank have facilitated capital flows between countries. The literature suggests that financial openness might be conducive to economic growth (see, for example, Kong (2021); Jamel and Maktouf (2017); Blanchard et al. (2017); Benigno et al. (2015)). Despite the perceived benefits, openness in terms of trade and finance has been increasingly treated with suspicion, partly due to the changing global geopolitics. For example, India balked at signing the Regional Comprehensive Economic Partnership (RCEP) out of fear that goods and services from other countries might destroy the Indian market (Palit 2021). Similarly, the US canceled its plan to sign the Trans-Pacific Partnership agreement, suspecting that further opening its financial and goods market to the participating countries would erode its competitiveness and eventually drag its economic growth (Chow et al. 2018). Given the increasing suspicions about economic and financial integration worldwide, it is incumbent to throw more light on this important topic.

In addition to the financial system, the literature has identified various factors that drive the effectiveness of policy. A strand of the recent literature focuses on the institutional theory of policy effectiveness. Specifically, the literature has focused mainly on how economics and financial institutions affect financial markets, income distribution, tax policies, and the effectiveness of policies formulated to strengthen the financial sector and the welfare of common people. Mishra et al. (2010) show that policy effectiveness can be enhanced in countries where autonomous institutions previously predicted central planning. Moreover, Cetorelli and Goldberg (2012) and Alpanda and Aysun (2012) provide evidence of a positive association between globalization and monetary policy effectiveness.

It is essential to understand that while many theoretical models, including the New Keynesian Models, have long recognized the possible role of financial development on fiscal and monetary policies, the empirical findings provide contradicting results. To achieve our goal, similar to several studies in the literature (e.g., Aysun and Hepp 2011; Ma and Lin 2016; Acemoglu et al. 2019), we also explore several options to estimate the relationship. While taking advantage of regression in a panel data setting, we employ the Pooled Regression, Random Effect, or Fixed Effect models. In addition to the above-mentioned models, we also utilize a Dynamic Panel Regression model, which addresses potential endogeneity problems in macroeconomic data (Arellano and Bond 1991).

We obtained several interesting results and observations. A summary of the important results is as follows. First, the results show that a higher level of financial development is associated with lower efficiency of monetary and fiscal policies regarding GDP growth. The relationship is significant only for monetary policy.

Second, this paper finds heterogeneity in the association between financial development and policy effectiveness. The diminishing impact of monetary policy on GDP growth is only significant in nations with a higher degree of financial development.

Third, the paper also sheds some light on the interplay between financial development and openness. Our results suggest that financial openness also decreases the efficacy of monetary policies regarding economic growth, and this association is economically significant. Additionally, financial development and openness display insignificant negative relationships.

Fourth, the negative association between policy effectiveness and financial development in terms of GDP growth is found to be independent of financial openness. This suggests that capital inflow or outflow from countries does not affect the dynamics between financial development and policies in terms of increasing output.

Fifth, we also study the international spillover effects of the US financial crisis and the role of policies and financial development during the crisis. The results show that the US financial crisis dampened the effectiveness of policy variables on GDP. However, financial development tends to partially neutralize the negative impacts of recessions.

In the rest of the study, Section 2 explains the proxies for financial development and openness variables. Section 3 details the process of empirical estimation, and Section 4 explains the results. Finally, Section 5 concludes the study.

2. Financial Development Index and Financial Openness Index

2.1. Financial Development Index (FDI)

An important strand of the financial economics literature has studied the impact of financial development in a nation on its economic growth (see, for example, Bernanke et al. 1999; Rajan and Zingales 1998; Levine 2005; Arcand et al. 2012; Durusu-Ciftci et al. 2017. In the literature, financial development has been defined in the following ways: (1) the ratio of domestic credit to GDP and (2) stock market capitalization as a percentage of GDP. Rajan and Zingales (1998) utilize both measures of financial development and conclude that economic growth is facilitated by financial development. On the other hand, Arcand et al. (2012) use only the credit-to-GDP ratio to prove that financial development ceases to impact economic growth positively below a certain threshold of economic development. Dabla-Norris and Srivisal (2013) also employ the credit-to-GDP ratio measure to show that financial development dampens the volatility of output, investment, and consumption to a certain extent.

With globalization and economic liberalization, financial development has undergone tremendous change worldwide. For example, various financial institutional actors, like mutual funds, pension funds, venture funds, co-operatives, etc., play significant roles in lending and borrowing with commercial banks. In addition, with the emergence of digital monies and other changes, corporate sectors can raise investments without regard for commercial banks and other established institutions. These changes require us to study the impact of financial development on economic growth using a multifaceted proxy of financial development. Therefore, we used the financial development proxy developed by Svirydzenka (2016). Svirydzenka (2016) developed a financial development proxy that comprises three major components calculated from different nationally and internationally known sources. To be specific, the proxy captures the financial market dept using the total value of stocks traded and private credit as a ratio of GDP obtained from the World Bank’s Global Financial Development Database. Similarly, access to financial services was calculated through the number of ATMs and bank accounts per capita from the IMF’s Financial Access Survey. Lastly, the efficiency of financial institutions was based on the net interest margin and overhead costs and sourced from the Bank for International Settlements. The authors then integrated these indicators using the principal component analysis and weighted averages, covering annual data from 1980 to 2018 for 183 countries. This structured methodology provides robust insights into the dynamics of financial markets and institutions, enhancing the credibility of our findings on financial development’s impact on economic growth. Figure 1 summarizes the construction process of the Financial Development Index (FDI) used in our study.

The FDI value ranges from 0 to 1. Figure 2 depicts the map of countries indicating the average FDI value from 1980 to 2018. The details in Figure 2 show that the USA, Australia, the UK, Canada, Japan, South Korea, and many European countries have higher values of FDI. Those countries’ high value of FDI represents highly developed financial systems, which are in accordance with the stylized information. Many emerging countries in the Asia-Pacific region, such as India, China, Brazil, South Africa, and the Russian Federation, are indicated by a white color representing the medium level of FDI. Lastly, countries in South Asia, Africa, Central Asia, and South America have lower index values, indicating lower levels of financial development across those countries.

2.2. Financial Openness Index (FOI)

We used the Chinn-Ito Index provided by Chinn and Ito (2006) as a proxy for financial openness. The Chinn-Ito index attempts to quantify restrictions on capital account transactions based on the IMF’s Annual Report on Exchange Arrangements and Exchange Restrictions (AREAER). During the process of index construction, dummy values are assigned to indicate the degree of capital control restrictiveness in categories such as equity, debt, and investment transactions. Once dummy variables are assigned, the financial openness index is constructed using a Principal Component Analysis of the dummy indicators. The higher the value of the index, the more open the country is. The Chinn-Ito index covers the period 1990–2018 and includes 182 countries, ensuring a diverse representation of global financial openness, which is illustrated in Figure 3.

Countries in a dark red color have higher FOI values, indicating less restriction on cross-border capital transactions. The figure shows that countries including the USA, Canada, the UK, Australia, Japan, Chile, and most European countries have higher financial openness. In comparison, Asia-Pacific countries like the Philippines, Indonesia, and the Russian Federation have a medium level of financial openness, with the index value ranging from 0.4 to 0.55. Lastly, most emerging countries, like India, China, Brazil, Vietnam, Central Asian countries, and South Africa, have lower index values ranging from 0 to 0.3, indicating restrictions in cross-border capital transactions.

Figure 2 and Figure 3 show that emerging countries like India, Brazil, South Africa, and China have a medium level of financial development but a lower level of financial openness. Therefore, studying the interaction between financial development and openness and their influence on policy effectiveness is imperative.

3. Empirical Methodology and Data

3.1. Empirical Methodology

The empirical framework assumes the following form:

$$\Delta y_{j,t}={\gamma }_o+{\Sigma }_{i=o}^p{\gamma }_i^y\Delta y_{j,t-i}+{\Sigma }_{i=o}^p{\gamma }_{j,t-i}^{po}\Delta p{o}_{j,t}+{\gamma }^{c}crisi{s}_{j,t}+u_j^y+w_{j,t}^y$$

(1)

In the equation, j refers to the countries and t is the time. Similarly, p represents the number of lags included in the regression estimate. $\Delta y$ is the output growth, and crisis is a dummy variable representing the crises in the US economy. $p{o}_t$ is a policy variable that represents fiscal or monetary policies. In the analysis, we used M1, the sum of cash currencies in circulation, and deposit money as a proxy of monetary policy, and tax revenue as an indicator of fiscal policy. In the robustness check, we also used M2, the sum of M1, and time and savings deposits with banks that require prior notice of withdrawal as a proxy of monetary policy. In the equation, $u_j^y$ is a country-fixed effect that captures the time-invariant characteristics of countries that could affect the growth of output and $w_{j,t}^y$ is the residual term that captures unobserved factors that could affect the growth of output in a country j in the year j. The rate of output growth, money growth rate, and tax revenue growth are defined as follows:

$$\Delta y_{j,t}=\frac{GD{P}_{j,t}-GD{P}_{j,t-1}}{GD{P}_{j,t-1}}$$

$$\Delta m_{j,t}=\frac{M{1}_{j,t}-M{1}_{j,t-1}}{M{1}_{j,t-1}}$$

$$\Delta T_{j,t}=\frac{T_{j,t}-T_{j,t-1}}{T_{j,t-1}}$$

It is evident from the first equation that the higher the coefficient of policy variable $p{o}_{j,t}$, the higher the policy effect on output growth. We now expand Equation (1) by incorporating the financial development variable into the regression framework. To study the role of financial development in conjunction with the policy variable, an interaction term is introduced as follows:

$${\gamma }_{j,t-i}^{po}={\theta }_i^{po}+{\theta }_i^{fdev}fde{v}_{j,t-i},$$

where $fde{v}_{j,t}$ is the financial development of country j at time t. ${\theta }_s$ are parameters. We can obtain an equation that explains the relationships between policy variables and financial development by incorporating Equations (1) and (2). The resultant equation is given as follows:

$$\begin{array}{cc}\hfill \Delta y_{j,t}=& {\gamma }_o+{\Sigma }_{i=o}^p{\gamma }_i^y\Delta y_{j,t-i}+\sum _{i=o}^p({\theta }_i^{po}\Delta p{o}_{j,t-i}+{\theta }_i^{fdev}fde{v}_{j,t-i}\hfill \\ & +{\theta }_i^{fpo}fde{v}_{j,t-i}\Delta p{o}_{j,t-i})+{\gamma }^{c}crisi{s}_{j,t}+u_j^y+w_{j,t}^y,\hfill \end{array}$$

(2)

where $fde{v}_{j,t}\Delta p{o}_{j,t}$ is an interaction term involving policy variables and financial development indicators. $crisi{s}_{j,t}$ is the indicator of US financial crises.

3.2. Data and Sources

This study utilizes several data sets from various sources. We used the Financial Development Index (FDI) from the IMF databases, the Financial Openness Index (FOI) from Chinn and Ito (2006) website, output data collected from World Bank databases, and data on Money Supply collected from the OECD website, while the policy variables were obtained from the World Development Indicators; lastly, the indicator $crisi{s}_t$ is the NBER recession indicator for the United States. The output growth is calculated as a percentage change in GDP per capita (constant 2010 USD) from the World Development Indicator. Considering the availability, the above-mentioned variables were collected annually for about 100 countries from 1980 to 2018.

4. Results

We can use three regression techniques to estimate panel data analysis: Fixed-Effect models, Pooled Least Squares models, and Random-Effects models. However, previous studies support using Fixed-Effects models because their results are more plausible under general market conditions and assumptions.

4.1. Baseline Results

This subsection presents the results under the baseline model framework—

Table 1. Financial Development and Policy Effectiveness: Baseline Results.

Dependent Variable: Real GDP Growth $\mathbf{\Delta }{\mathit{y}}_{\mathit{t}}$		Dependent Variable: Real GDP Growth $\mathbf{\Delta }{\mathit{y}}_{\mathit{t}}$
Independent Variables	Monetary Policy	Independent Variables	Fiscal Policy
Constant	1.673 *** (0.229)	Constant	1.645 *** (0.133)
$\Delta y_{t-1}$	0.477 *** (0.088)	$\Delta y_{t-1}$	0.44 *** (0.40)
$\Delta m_t$	0.011 *** (0.00467)	$\Delta T_t$	0.002 *** (0.00050)
$\Delta m_{t-1}$	$-0.001$ ** (0.001)	$\Delta T_{t-1}$	0.0003 (0.0004)
$\Delta m_{t-2}$	0.00004 (0.002)	$\Delta T_{t-2}$	0.0003 (0.0004)
$\Delta m_{t-3}$	−0.0004 ** (0.002)	$\Delta T_{t-3}$	0.0002 (0.0003)
$\Delta m_{t-4}$	$-0.00084$ (0.003)	$\Delta T_{t-4}$	0.0007 (0.0006)
$crisi{s}_t$	−1.8 *** (0.177)	$crisi{s}_t$	−1.52 *** (0.187)
${\sum }_{i=o}^4{\theta }_i^m$	0.0088	${\sum }_{i=o}^4{\theta }_i^T$	0.0035
Number of Observations	751	Number of Observations	2062
Number of Countries	27		100
F-statistic	196.11 ***	F-Statistic	314.78 ***
$R^2$	0.86	$R^2$	0.8737

${\sum }_{i=o}^4{\theta }_i^m$ is the sum of money coefficients $\Delta m_{t}s$ in the output growth equation. ${\sum }_{i=o}^4{\theta }_i^T$ is the sum of tax revenue coefficients $\Delta T_{ts}$ in the output growth equations. Values in parentheses are standard errors. *** means significant at 1%, ** at 5%, and * at 10%.

and

Table 2. Financial Development and Policy Effectiveness: Fixed Effects Results.

Dependent Variable: Real GDP Growth $\mathbf{\Delta }{\mathit{y}}_{\mathit{t}}$		Dependent Variable: Real GDP Growth $\mathbf{\Delta }{\mathit{y}}_{\mathit{t}}$
Independent Variables	Monetary Policy	Independent Variables	Fiscal Policy
Constant	1.749 *** (0.3156)	Constant	2.535 *** (0.4994)
$\Delta y_{t-1}$	0.274 *** (0.0645)	$\Delta y_{t-1}$	0.327 *** (0.502)
$\Delta m_t$	0.0012 *** (0.0004)	$\Delta T_t$	0.00114 ** (0.001)
$\Delta m_{t-1}$	−0.0006 (0.0004)	$\Delta T_{t-1}$	0.001 ** (0.0004)
$\Delta m_{t-2}$	−0.00001 (0.0002)	$\Delta T_{t-2}$	0.001 (0.001)
$\Delta m_{t-3}$	−0.0004 (0.0002)	$\Delta T_{t-3}$	0.00029 (0.004)
$\Delta m_{t-4}$	0.00003 (0.0002)	$\Delta T_{t-4}$	0.001 * (0.001)
$crisi{s}_t$	−4.9025 *** (0.8091)	$crisi{s}_t$	$-5.0694$ *** (0.7626)
${\sum }_{i=o}^4{\theta }_{i^m}$	0.00022	${\sum }_{i=o}^4{\theta }_{y^T}$	0.0039
Number of Observations	751	Number of Observations	2062
Number of Countries	27		100
F-statistic		F-Statistic
$R^2$	0.5299	$R^2$	0.4092

${\sum }_{i=o}^4{\theta }_i^m$ is the sum of money coefficients $\Delta m_{t}s$ in the output growth equation. ${\sum }_{i=o}^4{\theta }_i^T$ is the sum of tax revenue coefficients $\Delta T_{t}s$ in the output growth equations. Values in parentheses are standard errors. *** means significant at 1%, ** at 5%, and * at 10%.

present results under the Pooled Regression and the Fixed-Effects Models.

On the left side of Table 1, we display the effectiveness of monetary policies, while the outcomes of tax/fiscal policies are shown on the right. Table 1 shows that output growth persists yearly from 0.44 to 0.47 and is significant in both cases.

Regarding the monetary policy rule, one of the primary focuses of this study is the coefficient on money growth, which is significant and positive in the years of the policy’s enactment. However, the coefficients on lagged variables of money growth do not have a single significant sign. Also, the ${\sum }_{i=o}^4{\theta }_i^m$ has a calculated value of 0.0088, which is also non-significant. Theis finding suggests that monetary policy has a significan and positive effect on output growth in the year of policy formulation. However, the monetary policy does not significantly impact economic growth in the medium term. However, in the medium run, monetary policy does not play a substantial role in spurring economic growth. As expected, the coefficient of $crisis$ is significant and negative in both cases, suggesting that the US financial crisis dampens global output. Fiscal/tax policy is another primary focus of the study. The coefficients on tax revenues are positive but insignificant.

Regarding the model’s validity, the values $R^2$ in Table 1 are 0.86 and 0.87, representing the strong association between independent and dependent variables. Moreover, the relationship between output growth and policy variables does not change under the Fixed-Effects regression model. However, we find that US financial crises have adverse effects on the global economy under the Fixed-Effects and the Pooled Regression model and that the impacts are more extensive under the Fixed-Effects model than the Pooled Regression model. The results from the Fixed-Effects models are shown in Table 2.

4.2. Role of Financial Development

We intend to explore the role of financial development in the impact of monetary and fiscal policies on GDP growth. We do so by estimating Equation (2) under the Fixed-Effects model and using the panel data set of 99 countries from 1980 to 2018. By introducing financial development in the model, we observe a decrease in the sample size of countries due to a decrease in the degree of freedom. The results of the Fixed-Effects regression model are presented in

Table 3. Financial Development and Policy Effectiveness.

Dependent Variable: Real GDP Growth $\mathbf{\Delta }{\mathit{y}}_{\mathit{t}}$		Dependent Variable: Real GDP Growth $\mathbf{\Delta }{\mathit{y}}_{\mathit{t}}$
Independent Variables	Monetary Policy	Independent Variables	Fiscal Policy
Constant	2.48 * (1.33)	Constant	2.277 *** (0.4939)
$\Delta y_{t-1}$	0.247 *** (0.066)	$\Delta y_{t-1}$	0.30342 *** (0.04965)
$f{d}_t$	4.345 * (2.35)	$f{d}_t$	4.90 ** (2.29)
$fde{v}_{t-1}$	$-4.12$ * (4.42)	$fde{v}_{t-1}$	0.4868 (2.57)
$fde{v}_{t-2}$	2.70 (3.99)	$fde{v}_{t-2}$	−5.389 * (2.76)
$fde{v}_{t-3}$	0.321 * (3.42)	$fde{v}_{t-3}$	$-1.159$ (2.99)
$fde{v}_{t-4}$	$-4.53$ *** (1.87)	$fde{v}_{t-4}$	−3.0065 (2.7462)
$\Delta m_t$	0.0113 *** (0.0004)	$\Delta T_t$	0.0027 ** (0.00128)
$\Delta m_{t-1}$	0.00052 (0.011)	$\Delta T_{t-1}$	0.00076 (0.0012)
$\Delta m_{t-2}$	0.0007 (0.0008)	$\Delta T_{t-2}$	−0.0006 (0.0012)
$\Delta m_{t-3}$	−0.00146 (0.00089)	$\Delta T_{t-3}$	−0.0001 (0.00089)
$\Delta m_{t-4}$	0.00112 (0.001)	$\Delta T_{t-4}$	0.0003 (0.00180)
$fde{v}_t\Delta m_t$	0.00005 (0.0002)	$fde{v}_t\Delta T_t$	−0.0042 * (0.00256)
$fde{v}_{t-1}\Delta m_{t-1}$	−0.0018 (0.00137)	$fde{v}_{t-1}\Delta T_{t-1}$	0.0001 (0.0022)
$fde{v}_{t-2}\Delta m_{t-2}$	−0.0013 (0.0012)	$fde{v}_{t-2}\Delta T_{t-2}$	0.0027 (0.0022)
$fde{v}_{t-3}\Delta m_{t-3}$	0.001292	$fde{v}_{t-3}\Delta T_{t-3}$	0.0029 * (0.002)
$fde{v}_{t-4}\Delta m_{t-4}$	−0.0018 (0.0014)	$fde{v}_{t-4}\Delta T_{t-4}$	0.0015 (0.0021)
$crisi{s}_t$	−4.516 *** (1.246)	$crisi{s}_t$	−3.9 *** (0.8021)
${\sum }_{i=0}^4{\theta }_{i^{f}dev}$	−1.279	${\sum }_{i=0}^4{\theta }_{i^{f}dev}$	−8.582
${\sum }_{i=0}^4{\theta }_{i^m}$	0.00212	${\sum }_{i=0}^4{\theta }_{i^T}$	0.0021622
${\sum }_{i=0}^4{\theta }_{i^{f}m}$	−0.0054	${\sum }_{i=0}^4{\theta }_{i^{f}T}$	0.003123
Number of Observations	724	Number of Observations	2053
Number of Countries	27	Number of Countries	99
Adj-$R^2$	0.5409	Adj-$R^2$	0.4239

${\sum }_{i=o}^4{\theta }_i^m$ is the sum of money coefficients $\Delta m_{t}s$ in the output growth equation. ${\sum }_{i=o}^4{\theta }_i^T$ is the sum of tax revenue coefficients $\Delta T_{t}s$ in the output growth equations. ${\sum }_{i=0}^4{\theta }_{i,fdev}$ are sum of coefficients $fde{v}_{i,t}$. ${\sum }_{i=o}^4{\theta }_{i,fm}$ and ${\sum }_{i=o}^4{\theta }_{i,fT}$ are coefficients of the interaction terms $fde{v}_t\Delta m_t$ and $fde{v}_t\Delta T_t$. Values in parentheses are standard errors. *** means significant at 1%, ** at 5%, and * at 10%.

.

In Table 3, the effectiveness of monetary policies is displayed on the left, while the outcomes of fiscal/tax policies are shown on the right. The results show that economic output exhibits strong persistence, as indicated by the significant and positive Autoregressive Regression of order 1, AR(1), term in both specifications. We also observe a significantly negative coefficient of the $crisis$ variable, as indicated in Table 1 and Table 2. This finding again suggests that financial crises in the US economy have significant and adverse effects on the global economy. Consistent with the previous literature (see, for instance, Levine 1997; Khan 2001; Hassan et al. 2011; Ma and Lin (2016), financial developments positively impact economic growth in terms of monetary and fiscal policy specifications. However, financial development appears to have non-significant impacts on output growth in the medium run, as indicated by the positive and insignificant coefficients of ${\sum }_{i=o}^4{\theta }_i^{fdev}$.

When interacting with financial development, the monetary policy coefficient, $fde{v}_t\Delta m_t$, is much smaller than that of $\Delta m_t$. This result suggests that financial development lowers monetary policy’s usefulness in terms of economic growth, but this association is insignificant. In addition, the coefficient of ${\sum }_{i=o}^4{\theta }_i^{fm}$ is also negative and insignificant. The finding suggests that financial development negatively impacts the efficacy of monetary policy on output growth, but insignificantly, in the short-term and medium-term. When interacting with financial development, the coefficient of tax policy, $fde{v}_t\Delta T_t$, is negative and significant. This suggests that financial development blunts the marginal effectiveness of fiscal policy on output growth.

4.3. Financial Development and Heterogeneous Policy Effects on Output Growth

Now, we study whether financial development has heterogeneous policy effects on output growth across different countries. For this purpose, we categorized countries, based on the average level of financial development calculated over the sample period, into three quantiles. Due to the lack of data on monetary policy variables for the countries in the second quantile, the analysis focused only on the first and third quantiles.

Table 4. Financial Development and Heterogeneous Policy Effect.

Dependent Variable: Real GDP Growth $\mathbf{\Delta }{\mathit{y}}_{\mathit{t}}$			Dependent Variable: Real GDP Growth $\mathbf{\Delta }{\mathit{y}}_{\mathit{t}}$
Monetary Policy			Fiscal Policy
Independent Variables	LFD	HFD	Independent Variables	HFD	LFD
Constant	1.78 *** (0.50)	2.198 *** (0.251)	Constant	1.67 *** (0.165)	2.099 *** (0.239)
$\Delta y_{t-1}$	0.10 (0.087)	0.341 *** (0.033)	$\Delta y_{t-1}$	0.386 *** (0.041)	0.285 *** (0.509)
$\Delta m_t$	0.003 (0.002)	0.005 *** (0.0014)	$\Delta T_t$	0.0004 (0.0022)	0.0017 (0.0035)
$\Delta m_{t-1}$	0.003 (0.002)	−0.0043 ** (0.002)	$\Delta T_{t-1}$	0.0003 (0.0029)	−0.0014 (0.0048)
$fde{v}_t\Delta m_t$	−0.0009 (0.0045)	−0.0073 *** (0.002)	$fde{v}_t\Delta T_t$	0.0026 (0.0035)	0.0021 (0.01)
$fde{v}_{t-1}\Delta m_{t-1}$	$-0.011$ * (0.054)	0.0049 ** (0.0025)	$fde{v}_{t-1}\Delta T_{t-1}$	$-0.000233$ (0.005)	0.0077 (0.014)
$crisi{s}_t$	$-1.582$ ** (0.6638)	$-1.598$ *** (0.195)	$crisi{s}_t$	$-1.44$ *** (0.175)	$-1.499$ *** (0.485)
${\sum }_{i=0}^1{\theta }_{i^m}$	0.006	0.0007	${\sum }_{i=0}^4{\theta }_{i^T}$	0.0003	0.0003
${\sum }_{i=0}^1{\theta }_{i^{f}m}$	$-0.0119$	$-0.0024$	${\sum }_{i=0}^4{\theta }_{i^{f}T}$	0.00277	0.0098
Number of Observations	158	641	Number of Observations	1259	656
Number of Instruments	159	552	Number of Instruments	672	596
Number of Countries	6	21	Number of Countries	42	32
Wald ${\chi }^2$	181.03 ***	340.73 ***	Wald ${\chi }^2$	275.95 ***	33.92 ***

${\sum }_{i=o}^1{\theta }_i^m$ is the sum of money coefficients $\Delta m_{t}s$ in the output growth equation. ${\sum }_{i=o}^1{\theta }_i^T$ is the sum of tax revenue coefficients $\Delta T_{t}s$ in the output growth equations. ${\sum }_{i=o}^1{\theta }_{i,fm}$ and ${\sum }_{i=o}^4{\theta }_{i,fT}$ are coefficients of the interaction terms $fde{v}_t\Delta m_t$ and $fde{v}_t\Delta T_t$. Values in parentheses are standard errors. *** means significant at 1%, ** at 5%, and * at 10%. HFD refers to the highly financially developed countries, and LFD refers to the countries with low levels of financial development. Countries with higher levels of financial development are those in North America, Western Europe, and East Asia, while countries in South Asia, Central Asia, East Europe, and Africa were calculated to have low levels of financial development index.

presents the results of this exercise.

Our results show that financial development diminishes the efficacy of monetary policy in countries with higher and lower levels of financial development. For example, the coefficients indicating the impact of monetary policy in Less Financially Developed Countries and Highly Financially Developed Countries are 0.003 and 0.005, respectively. In addition, the table also suggests that an increase in monetary aggregates has a significant effect in Highly Financially Developed Countries. However, the diminishing effects of financial development on monetary policy efficacy are more pronounced and significant in countries with higher levels of financial development, as concluded from the coefficients of the interaction terms, $fde{v}_t\Delta m_t$, $-0.0074$ **. The opposite is true for countries with lower levels of financial development. In the case of fiscal policy, financial development impacts output growth positively, although non-significantly. Table 4 also suggests that financial crises in the US economy negatively affect highly developed and less developed countries.

4.4. Financial Development and Policy Effectiveness: A Dynamic GMM Approach

As proposed by Arellano and Bond (1991) and Blundell and Bond (1998), to further check robustness, we conducted our benchmark model analysis using the System Generalized Method of Moments (SYS-GMM). This method also helped us address the endogeneity problem that arises from the use of the lag terms of our independent variables. We present the results in

Table 5. Financial Development and Policy Effectiveness: SYS-GMM Approach.

Dependent Variable: Real GDP Growth $\mathbf{\Delta }{\mathit{y}}_{\mathit{t}}$		Dependent Variable: Real GDP Growth $\mathbf{\Delta }{\mathit{y}}_{\mathit{t}}$
Independent Variables	Monetary Policy	Independent Variables	Fiscal Policy
Constant	2.045 * (0.756)	Constant	3.669 *** (0.8)
$\Delta y_{t-1}$	0.22 *** (0.054)	$\Delta y_{t-1}$	0.238 * (0.052)
$\Delta m_t$	0.0054 *** (0.001)	$\Delta T_t$	0.004 * (0.002)
$\Delta m_{t-1}$	$-0.0026$ ** (0.0012)	$\Delta T_{t-1}$	0.0013 (0.002)
$\Delta m_{t-2}$	−0.0001 (0.0008)	$\Delta T_{t-2}$	0.001 (0.001)
$\Delta m_{t-3}$	−0.0005 (0.0008)	$\Delta T_{t-3}$	−0.001 (0.001)
$\Delta m_{t-4}$	−0.0003 (0.0009)	$\Delta T_{t-4}$	0.0005 (0.002)
$fde{v}_t$	9.65 *** (2.84)	$fde{v}_t$	6.81 *** (2.65)
$fde{v}_{t-1}$	−3.45 (5.30)	$fde{v}_{t-1}$	2.26 (2.86)
$fde{v}_{t-2}$	$-6.21$ ** (2.86)	$fde{v}_{t-2}$	$-9.23$ *** (3.035)
$fde{v}_{t-3}$	7.249(3.41)	$fde{v}_{t-3}$	2.723 (3.06)
$fde{v}_{t-4}$	$-7.02$ *** (2.10)	$fde{v}_{t-4}$	$-6.22$ *** (2.87)
$fde{v}_t\Delta m_t$	$-0.0072$ *** (0.0014)	$fde{v}_t\Delta T_t$	−0.002 (0.002)
$fde{v}_{t-1}\Delta m_{t-1}$	$0.0031$ * (0.02)	$fde{v}_{t-1}\Delta T_{t-1}$	0.0005 (0.003)
$fde{v}_{t-2}\Delta m_{t-2}$	−0.00002 (0.0011)	$fde{v}_{t-2}\Delta T_{t-2}$	0.0006 (0.00235)
$fde{v}_{t-3}\Delta m_{t-3}$	0.00012 (0.0011)	$fde{v}_{t-3}\Delta T_{t-3}$	0.004 * (0.0021)
$fde{v}_{t-4}\Delta m_{t-4}$	−0.0001 (0.0012)	$fde{v}_{t-4}\Delta T_{t-4}$	0.00141 (0.00284)
$crisi{s}_t$	−1.577 *** (2.10)	$crisi{s}_t$	$-1.51$ *** (0.215)
${\sum }_{i=0}^1{\theta }_{i^{f}dev}$	0.209	${\sum }_{i=0}^1{\theta }_{i^{f}dev}$	−3.6
${\sum }_{i=0}^1{\theta }_{i^m}$	0.0018	${\sum }_{i=0}^4{\theta }_{i^T}$	0.005
${\sum }_{i=0}^1{\theta }_{i^{f}m}$	$-0.004$	${\sum }_{i=0}^4{\theta }_{i^{f}T}$	0.0039
Number of Observations	724	Number of Observations	1928
Number of Instruments	677	Number of Instruments	677
Number of Countries	27	Number of Countries	98
Wald ${\chi }^2$	1178.43 ***	Wald ${\chi }^2$	253.82 ***

${\sum }_{i=o}^1{\theta }_i^m$ is the sum of money coefficients $\Delta m_{t}s$ in the output growth equation. ${\sum }_{i=o}^1{\theta }_i^T$ is the sum of tax revenue coefficients $\Delta T_{t}s$ in the output growth equations. ${\sum }_{i=o}^1{\theta }_{i,fm}$ and ${\sum }_{i=o}^4{\theta }_{i,fT}$ are coefficients of the interaction terms $fde{v}_t\Delta m_t$ and $fde{v}_t\Delta T_t$. Values in parentheses are standard errors. *** means significant at 1%, ** at 5%, and * at 10%.

. The results in the Table suggest that the monetary policy variable, $\Delta m_t$, has positive and significant impacts on output growth. To be specific, the GMM model suggests that an increase in money supply by a percent increases output growth by 0.0054%. However, the positive effects of monetary policy are reduced in the presence of financial development, given by the coefficient of $0.0072$ *** on the interaction term, $fde{v}_t\Delta m_t$. This result reaffirms that a nation’s financial development significantly diminishes the efficacy of monetary policy regarding economic output. In the context of fiscal policy, the table suggests that financial development is not likely to play a significant role in tax policies’ impacts on output growth. Additionally, the table suggests that crises in the US significantly impact the global economy, with a negative coefficient of $1.577$ ***. The above findings from the dynamic GMM approach are similar to those obtained in the baseline model, affirming that the relationship between policy variables and financial development is robust across different econometric methods. Furthermore, the major outcomes from the baseline models are preserved even when we use the three-year moving averages of the variables during the estimation process. These results are presented in

Table 6. Financial Development and Policy Effectiveness: 3-year Moving Averages.

Dependent Variable: Real GDP Growth $\mathbf{\Delta }{\mathit{y}}_{\mathit{t}}$		Dependent Variable: Real GDP Growth $\mathbf{\Delta }{\mathit{y}}_{\mathit{t}}$
Independent Variables	Monetary Policy	Independent Variables	Fiscal Policy
Constant	1.147 * (0.6543)	Constant	1.2036 *** (0.251)
$\Delta y_{t-1}$	0.773 *** (0.02256)	$\Delta y_{t-1}$	0.772 *** (0.0177)
$\Delta m_t$	0.004 *** (0.0011)	$\Delta T_t$	0.0039 ** (0.0013)
$\Delta m_{t-1}$	$-0.0037$ *** (0.0013)	$\Delta T_{t-1}$	−0.0019 (0.0015)
$f{d}_t\Delta m_t$	$-0.0046$ *** (0.00155)	$f{d}_t\Delta T_t$	$-0.006$ ** (0.0025)
$f{d}_{t-1}\Delta m_{t-1}$	$0.0038$ ** (0.0018)	$f{d}_{t-1}\Delta T_{t-1}$	0.0048 * (0.0028)
$crisi{s}_t$	$-0.4464$ (0.5465)	$crisi{s}_t$	$-0.898$ *** (0.3028)
${\sum }_{i=0}^1{\theta }_{i^m}$	0.0003 ***	${\sum }_{i=0}^1{\theta }_{i^T}$	0.002 **
${\sum }_{i=0}^1{\theta }_{i^{f}m}$	$-0.0008$ ***	${\sum }_{i=0}^1{\theta }_{i^{f}T}$	$-0.0012$ **
Number of Observations	778	Number of Observations	2167
Number of Countries	27	Number of Countries	102
Adj-$R^2$	0.8502	Adj-$R^2$	0.7815

${\sum }_{i=o}^1{\theta }_i^m$ is the sum of money coefficients $\Delta m_{t}s$ in the output growth equation. ${\sum }_{i=o}^1{\theta }_i^T$ is the sum of tax revenue coefficients $\Delta T_{t}s$ in the output growth equations. ${\sum }_{i=o}^1{\theta }_{i,fm}$ and ${\sum }_{i=o}^4{\theta }_{i,fT}$ are coefficients of the interaction terms $fde{v}_t\Delta m_t$ and $fde{v}_t\Delta T_t$. Values in parentheses are standard errors. *** means significant at 1%, ** at 5%, and * at 10%.

. Similar to the previous finding, Table 6 suggests that monetary and fiscal policies tend to have positive and significant impacts on output growth. However, the positive policies’ impacts on output growth are diminished when interacting with financial development. The decrease is significant in the context of monetary policy but also in the case of fiscal policy. The significant diminishing effect of fiscal policy when interacting with financial development ($-0.006$ **) suggests that financial development might interact with fiscal policy in the medium term rather than in the short term. Nevertheless, these findings once again affirm that financial development decreases policy effectiveness regarding output growth.

4.5. Financial Openness, Financial Development, and Policy Effectiveness

In this section, we explore the role of financial openness in the connection between financial development and policy effectiveness. A few studies that we found present inclusive results regarding financial openness, financial development, and policy effectiveness. By using the Quinn Index as a proxy of financial openness, Edwards (2001) concludes that a nation must have a strong and developed domestic financial system to reap the benefits of financial openness. Nicolo and Juvenal (2010) discuss that increasing financial integration among nations impacted financial development and economic growth positively between 1980 and 2009. Still, many doubt the potential of financial openness at the political level. Therefore, we attempt to evaluate the role of financial openness in the dynamics between policy effectiveness and economic growth. We are also interested in evaluating the role of financial openness in the relationship between financial development and policy variables. We follow Dawson and Richter (2006), and employ a three-way interaction in panel regression as follows:

$$\begin{array}{cc}\hfill \Delta y_{j,t}=& {\gamma }_o+{\Sigma }_{i=o}^p{\gamma }_i^y\Delta y_{j,t-i}+\sum _{i=o}^p({\theta }_i^{po}\Delta p{o}_{j,t-i}+{\theta }_i^{fdev}fde{v}_{j,t-i}+{\theta }_i^{fpo}fde{v}_{j,t-i}\Delta p{o}_{j,t-i}\hfill \\ & +{\theta }_i^{fopen}fope{n}_{j,t-i}+{\theta }_i^{fofd}fde{v}_{j,t-i}fope{n}_{j,t-i}+{\theta }_i^{fopo}fope{n}_{j,t-i}\Delta p{o}_{j,t-i}\hfill \\ & +{\theta }_i^{fofdpo}fope{n}_{j,t-i}fde{v}_{j,t-i}\Delta p{o}_{j,t-i})+{\gamma }^{c}crisi{s}_{j,t}+u_{j,t}^y,\hfill \end{array}$$

(3)

where $fde{v}_{j,t}$, $p{o}_{j,t}$, and $crisi{s}_{j,t}$ have usual meaning. The variable $fope{n}_{j,t}$ is the financial openness of a country j in year t. $fope{n}_{j,t}fde{v}_{j,t}$ is the interaction term between financial development and financial openness. Similarly, $fope{n}_{j,t}\Delta p{o}_{j,t}$ is an interaction term regarding financial openness and policy variables. Lastly, $fope{n}_{j,t}fde{v}_{j,t}\Delta p{o}_{j,t}$ is the three-way interaction between financial openness, financial development, and policy variables. We present the estimation results of Equation (3) in

Table 7. Financial Development, Financial Openness, and Policy Effectiveness.

Financial Development, Financial Openness and Policy Effectiveness		Financial Development, Financial Openness, and Policy Effectiveness
Dependent Variable: Real GDP Growth$\Delta {\mathbf{y}}_{\mathbf{t}}$		Dependent Variable: Real GDP Growth $\Delta {\mathbf{y}}_{\mathbf{t}}$
Independent Variables	Monetary Policy	Independent Variables	Fiscal Policy
Constant	1.44 * (0.80)	Constant	1.44 *** (0.220)
$\Delta y_{t-1}$	0.2411 *** (0.058)	$\Delta y_{t-1}$	0.310 *** (0.04)
$\Delta m_t$	0.0009 * (0.0022)	$\Delta T_t$	0.0017 * (0.0016)
$\Delta m_{t-1}$	0.004 * (0.002)	$\Delta T_{t-1}$	0.00067 (0.0012)
$Opennes{s}_t$	0.069 * (0.763)	$Opennes{s}_t$	1.31 *** (3.66)
$fde{v}_t\Delta m_t$	−0.093 ** (0.0044)	$fde{v}_t\Delta T_t$	−0.0017 (0.0052)
$fde{v}_{t}fopen{n}_t$	−0.80 (1.40)	$fde{v}_{t}fope{n}_t$	$-3.94$ *** (0.95)
$\Delta m_{t}fope{n}_t$	$-0.0047$ ** (0.002)	$\Delta T_{t}fope{n}_t$	0.0013 (0.0023)
$\Delta m_{t}fope{n}_{t}f{d}_t$	−0.0048 (0.0048)	$\Delta T_{t}Opennes{s}_{t}f{d}_t$	−0.0032 (0.0058)
$crisi{s}_t$	$-1.44$ * (0.939)	$crisi{s}_t$	$-3.030$ *** (0.80455)
Number of Observations	832	Number of Observations	2438
Number of Countries	27	Number of Countries	107
$R^2$	0.54	$R^2$	0.382

Values in parenthesis are standard errors. *** means significant at 1%, ** at 5%, and * at 10%.

.

Our results in Table 7 show that financial openness and monetary and fiscal policies positively and significantly impact output growth. As in our baseline model, the interaction between financial development and monetary policy is negatively significant, which confirms that financial development dampens the efficacy of monetary policy on output growth. Moreover, the interaction term between financial development and openness variables has a negative but insignificant coefficient. However, the interaction between financial openness and monetary policy is negative but significant, suggesting that financial openness diminishes the impact of monetary policy on GDP growth. The same holds for the Fiscal Policy variable. The three-way interaction between financial development, financial openness, and the policy variable has negative and insignificant coefficients. The negative coefficients indicate that financial openness non-significantly diminishes the interaction between financial development and monetary (fiscal) policies in influencing output growth. Again, we confirm the negative and significant impact of the US financial crisis on the global economy.

4.6. Financial Development, Policy Effectiveness, and the State of the US Economy

In this subsection, we study the international spillover effects of the US financial crises and the impact that the financial development of a host country has on these effects. To do so, we re-estimated the GMM model with a new interaction term indicating crises in the US economy. The modified GMM model framework for estimation is given as follows:

$$\begin{array}{cc}\hfill \Delta y_{j,t}=& {\gamma }_o+{\Sigma }_{i=o}^p{\gamma }_i^y\Delta y_{j,t-i}+\sum _{i=o}^p({\theta }_i^{po}\Delta p{o}_{j,t-i}+{\theta }_i^{fdev}fde{v}_{j,t-i}\hfill \\ & +{\theta }_i^{fdcrisis}fde{v}_{j,t-i}crisi{s}_{t-i}+{\theta }_i^{crisispo}p{o}_{j,t-i}crisi{s}_{t-i}+{\theta }_i^{fdevpo}fde{v}_{j,t-i}\Delta p{o}_{j,t-i}\hfill \\ & +{\theta }_i^{focrisispo}crisi{s}_{t-i}fde{v}_{j,t-i}\Delta p{o}_{j,t-i})+{\gamma }^{c}crisi{s}_t+u_{j,t}^y,\hfill \end{array}$$

(4)

where $fde{v}_{j,t}$, $p{o}_{j,t}$, and $crisi{s}_{j,t}$ have their usual meaning. $crisi{s}_{t}fde{v}_{j,t}$ is the interaction term between financial development and US economic crises. Similarly, $crisi{s}_t\Delta p{o}_{j,t}$ is an interaction term between US financial crises and policy variables. Lastly, $crisi{s}_{t}fde{v}_{j,t}\Delta p{o}_{j,t}$ is the three-way interaction between US crises, financial development, and policy variables.

We present Equation (4)’s estimation results in

Table 8. Spillover Effects of US Financial Crisis, Policy Effectiveness, and Financial Development.

Dependent Variable: Real GDP Growth ($\mathbf{\Delta }{\mathit{y}}_{\mathit{t}}$)		Dependent Variable: Real GDP Growth ($\mathbf{\Delta }{\mathit{y}}_{\mathit{t}}$)
Independent Variables	Monetary Policy	Independent Variables	Fiscal Policy
Constant	1.78 *** (0.77)	Constant	2.96 *** (0.76)
$\Delta y_{t-1}$	0.2 *** (0.05)	$\Delta y_{t-1}$	0.244 *** (0.048)
$\Delta m_t$	0.004 *** (0.001)	$\Delta T_t$	0.003 (0.002)
$\Delta m_{t-1}$	$-0.001$ * (0.0004)	$\Delta T_{t-1}$	0.001 * (0.001)
$\Delta m_{t-2}$	$-0.001$ (0.0002)	$\Delta T_{t-2}$	0.001 (0.0006)
$\Delta m_{t-3}$	$-0.0005$ *** (0.0001)	$\Delta T_{t-3}$	0.0004 (0.0005)
$\Delta m_{t-4}$	$-0.0004$ (0.0003)	$\Delta T_{t-4}$	0.001 (0.001)
$fde{v}_t$	9.19 *** (2.91)	$fde{v}_t$	7.95 *** (2.45)
$fde{v}_{t-1}$	$-2.66$ (5.2)	$fde{v}_{t-1}$	2.2 (2.8)
$fde{v}_{t-2}$	$-5.96$ ** (2.6)	$fde{v}_{t-2}$	$-9.26$ *** (3.1)
$fde{v}_{t-3}$	7.64 ** (2.6)	$fde{v}_{t-3}$	3.25 (2.95)
$fde{v}_{t-4}$	$-7.38$ *** (2.022)	$fde{v}_{t-4}$	$-6.44$ ** (2.5)
$crisi{s}_t$	$-1.30$ * (0.71)	$crisi{s}_t$	$-1.22$ ** (0.55)
$fde{v}_t\ast crisis$	$-0.77\left(1.1\right)$	$fde{v}_t\ast crisis$	−0.736 (0.95)
$fde{v}_t\ast \Delta m_t$	$-0.005$ *** (0.001)	$fde{v}_t\ast \Delta T_t$	−0.004 (0.004)
$\Delta m_t\ast crisis$	0.002 (0.002)	$\Delta T_t\ast crisis$	0.007 ** (0.0038)
$fde{v}_t\ast \Delta m_t\ast crisis$	−0.003 (0.002)	$fde{v}_t\ast T_t\ast crisis$	−0.01 (0.01)
${\sum }_{i=o}^4{\theta }_{i,\Delta m_t}$	$-0.001$	${\sum }_{i=o}^4{\theta }_{i,\Delta T_t}$	0.006
${\sum }_{i=o}^4{\theta }_{i,fde{v}_t}$	0.82	${\sum }_{i=o}^4{\theta }_{i,fde{v}_t}$	−2.4
Number of Observations	724	Number of Observations	1935
Number of Instruments	631	Number of Instruments	676
Number of Countries	27	Number of Countries	98
Wald ${\chi }^2$	900.59 ***	Wald ${\chi }^2$	222.32 ***

Notes: Values in parenthesis are standard errors. *** means significant at 1%, ** at 5%, and * at 10%.

. Our results show that monetary policy has a significantly positive impact on output growth. We find several similar results as those obtained in our baseline models; for example, the interaction between financial development and monetary policy has a statistically significant negative coefficient, reaffirming that the efficacy of monetary policy on output growth diminishes as countries become more financially developed.

However, we are interested in the behavior of crisis terms. Our results show that the coefficient of crisis terms is negative but significant, once again confirming that the US financial crisis negatively impacted global GDP growth. More interestingly, we find that a nation with strong financial development is not likely to mitigate the adverse spillover effects of the US financial crisis on global GDP, as indicated by the negative but insignificant coefficient of crisis and financial development terms. However, the coefficient of crisis (−1.3) is larger than that of the interaction term (−0.77). The difference suggests that strong financial development in a host country will likely partially negate the adverse impacts of the US financial crisis. However, the interaction between the two variables is not significant.

While research on the usefulness of monetary policy during a crisis is gaining attention, we provide some crucial results on this topic. Our results indicate that while monetary policy positively and significantly impacts GDP growth, this impact diminishes slightly during a crisis. This finding suggests that the efficacy of monetary policy can not stimulate the economy as efficiently during crises as it can during normal times. Lastly, the coefficient on the three-way interaction between financial development, monetary policy, and crisis is −0.003, which suggests that even during the crisis, financial development diminishes the efficacy of monetary policy in stimulating GDP growth. Since the coefficient is smaller than the interaction term without the crisis, we can say that the diminishing role of financial development in monetary policy efficacy decreases due to the crisis.

On the right side of the table, we can see the efficacy of fiscal policy under the modified empirical specifications. The table shows that tax revenue’s contribution to GDP growth during crisis times is larger than that during normal times. The table also suggests that financial development’s diminishing role in fiscal policy success decreases during the crisis period, as in the case of monetary policy.

4.7. Robustness Checks

Here, we change the proxy for financial development and instead use domestic bank credit issued as a percentage of GDP. The results of this robustness exercise are presented in

Table 9. Robustness Analysis.

Dependent Variable: Real GDP Growth ($\mathbf{\Delta }{\mathit{y}}_{\mathit{t}}$)		Dependent Variable: Real GDP Growth ($\mathbf{\Delta }{\mathit{y}}_{\mathit{t}}$)
Independent Variables	Monetary Policy	Independent Variables	Fiscal Policy
Constant	2.84 *** (0.481)	Constant	4.17 *** (0.785)
$\Delta y_{t-1}$	0.22 *** (0.059)	$\Delta y_{t-1}$	0.16 *** (0.049)
$\Delta m_t$	0.002 * (0.001)	$\Delta T_t$	$0.002$ * (0.001)
$\Delta m_{t-1}$	$-0.0007$ (0.0008)	$\Delta T_{t-1}$	0.001 (0.001)
$\Delta m_{t-2}$	0.0001 (0.0001)	$\Delta T_{t-2}$	0.005 (0.001)
$\Delta m_{t-3}$	−0.001 (0.0005)	$\Delta T_{t-3}$	$-0.0004$ (0.0009)
$\Delta m_{t-4}$	$-0.001$ * (0.0004)	$\Delta T_{t-4}$	0.0004 (0.0013)
$Credi{t}_t$	−0.002 (0.01)	$Credi{t}_t$	$-0.044$ *** (0.0168)
$Credi{t}_{t-1}$	0.014 (0.0227)	$Credi{t}_{t-1}$	0.021 (0.018)
$Credi{t}_{t-2}$	−0.008 (0.009)	$Credi{t}_{t-2}$	0.001 (0.01)
$Credi{t}_{t-3}$	−0.02 (0.014)	$Credi{t}_{t-3}$	$-0.04$ *** (0.0194)
$Credi{t}_{t-4}$	0.007 (0.009)	$Credi{t}_{t-4}$	$0.0284$ *** (0.0095)
$crisi{s}_t$	$-1.54$ *** (0.214)	$crisi{s}_t$	$-1.4$ *** (0.25)
$Credi{t}_t\ast \Delta m_t$	$-0.00001$ (0.00001)	$Credi{t}_t\ast \Delta T_t$	0.000088 (0.0000092)
Number of Observations	689	Number of Observations	1629
Number of Instruments	613	Number of Instruments	677
Number of Countries	27	Number of Countries	98
Wald ${\chi }^2$	1111.43 ***	Wald ${\chi }^2$	214.97 ***

Notes: Values in parenthesis are standard errors. *** means significant at 1%, ** at 5%, and * at 10%.

. Similar to the results of previous exercises, we find that the degree of financial development dampens the efficacy of monetary and fiscal policies in increasing a nation’s economic output. Hence, our findings are robust regardless of the financial development measures that we use.

5. Discussions, Conclusions, and Policy Implications

The literature on the interdependence between a nation’s financial development and policy effectiveness is gaining in popularity of late. While this is the case, the results from these studies are still mixed. This study adds to the literature by providing novel evidence of the association between financial development and policy effectiveness using a new proxy for financial development. We use panel data from 100 countries spanning from 1980 to 2018 and find that the degree of financial development adversely impacts the efficacy of monetary policy. Interestingly, the adverse effect of financial development on the effectiveness of monetary policy is more evident in countries with higher degrees of financial development at present. We find similar results when we analyze the marginal impact of financial development on the efficacy of fiscal/tax policies. Moreover, the findings also show that a financial crisis in the US economy dampens the effectiveness of policy in other countries, and that strong financial markets and institutions in these countries partially offset the negative impacts of the crisis.

In the case of financial openness, we find that this variable also dampens the effectiveness of monetary policy. However, we did not find any significant association between financial development and openness. Consequently, the degree of financial openness also does not meaningfully impact the interaction between financial development and policy effectiveness. Our results contribute to the literature by investigating the three-way dynamics of financial development, openness, and policy effectiveness.

There are several possible reasons for the observed association between financial development, openness, and policy success. For example, financially developed countries enjoy more stable and efficient financial intermediaries, less capital outflow (flight) due to political stability, and more instruments that affect GDP growth compared to countries that are financially less developed. Moreover, downturns in the US economy change people’s future expectations of policy variables and institutions. These changes in expectations tend to minimize the impact of policy variables during the crisis.

The findings in this paper have several policy implications. Our results suggest that the adverse effects of financial development on policy effectiveness are higher in developed countries compared to less developed countries. As such, rich countries should consider alternative tools to enhance policy effectiveness in normal (crisis-free) times. In addition, although the impact of financial development on the effectiveness of fiscal policy appears to be less pronounced compared to that of monetary policy, the impact is still significant. This suggests that financial development can blunt the efficacy of fiscal interventions, indicating the need for tailored fiscal policies that consider the financial maturity of the economy. Furthermore, our results highlight the role of robust financial systems in mitigating adverse effects from external shocks, such as those from US economic downturns. This points to the importance of resilient financial institutions and systems that could serve as buffers against external economic volatility.

Although our paper presents significant findings in terms of the association between financial development, financial openness, and policy effectiveness, they lack a robust theoretical framework. To better understand the relationship between these three variables, we plan to further this analysis in a micro-founded model that incorporates the interplay of financial development, openness, and policy alternatives in the near future.