Financial development plays a crucial role in shaping economic growth, yet it can ­introduce volatility. This study examines the relationship between financial development and economic growth volatility. Using panel data from 60 countries (30 de­veloped and 30 developing) for 1981–2022, we employ panel-corrected standard errors and generalized­ method of moments to ensure robustness. Financial development is analyzed through financial institutions and financial markets across three dimensions: depth, ­access, and ­efficiency. Conceptually, the paper finds that the supply-leading hypothesis does not account for the economic growth volatility associated with excessive financialization. The results indicate that, at higher levels, financial development has a volatility-enhancing impact in developed countries, while in developing countries it has a volatility-reducing effect. Policymakers in developed countries should ensure that credit expansion is aligned with real-sector development. Regulators should monitor adverse effects of financial depth and ensure funds are directed toward real-sector growth, while improving access and efficiency. In a too‑much-finance scenario, economies need moderators — such as strong regulatory quality and well-defined rights for creditors and borrowers — to mitigate volatility-enhancing effects.

financialization, financial development, economic growth volatility, financial depth, economic growth

Financial development (FD) encompasses the factors, policies, and institutions that contribute to effective financial intermediation and well-functioning markets, ensuring broad and inclusive access to capital and financial services (World Economic Forum, 2011). Conceptually, FD is a process aimed at reducing the costs of acquiring information, enforcing contracts, and conducting transactions (World Bank, 2019). The supply-leading hypothesis emerges from this context, positing that FD drives economic growth by ensuring the efficient allocation of financial resources.

As financial systems expand and persist over time, expectations and dependencies on intermediation increase, adding layers of complexity and accumulating systemic vulnerabilities. These fragilities may trigger a chain reaction of unexpected economic disruptions, potentially culminating in financial crises. Rajan (2005) was among the earliest researchers to systematically discuss the detrimental effects of excessive finance. A growing body of literature suggests a trade-off between economic growth and growth volatility. While many scholars argue that FD promotes long-term expansion, it may simultaneously contribute to financial instability and macroeconomic volatility. Key drivers include high inflation that affects consumption patterns, reduced investment, declining borrower net worth, systemic risks leading to banking crises, and financial liberalization that diverts human capital from productive sectors.

Economic growth volatility is commonly classified into two categories: realized volatility — measured as the standard deviation of per-capita GDP growth — and innovation volatility — defined as the standard deviation of unexpected growth shocks. According to Andersen and Banzoni (2008), realized volatility is a nonparametric ex-post estimate of the return variation. The most obvious realized volatility measure is the sum of finely-sampled squared return realizations over a fixed time interval. In contrast, innovation volatility may be referred to as the variation (over time) in aggregate economic output or other macroeconomic variables that arises due to structural, financial, or institutional innovations (e.g., financial-market development, financial liberalization, structural reforms, technological–financial innovation; Jermann and Quadrini, 2006). In this study, economic growth volatility (hereafter, EV) refers to realized­ volatility, assessed using the standard deviation of per-capita GDP growth.

A widely held view suggests that FD reduces macroeconomic volatility, particularly in developed countries. Some studies propose that well-developed banking systems mitigate EV, as industrial output fluctuations tend to be negatively correlated with banking-sector credit portfolios. However, these conclusions have not been consistently robust across different measures of FD. Given the crucial role that FD plays in shaping economic growth, policymakers must assess its implications for both stability and volatility. A precise understanding of this relationship is essential for designing macroeconomic policies that maximize the benefits of FD while mitigating its potential risks.

This relationship requires systematic evaluation from several perspectives. First, a simultaneous study of developed and developing countries using the same period, variable measures, and estimation techniques can clarify potentially different dynamics. Second, in contrast to earlier unidimensional approaches, this study examines depth, access, and efficiency for both bank-based and market-based components of FD.

We address three questions. First, does excessive FD prove detrimental by inducing EV? Second, if so, which dimensions of FD (depth, access, or efficiency) are the primary contributors? Third, is this effect consistent across developed and developing countries? The answers are intended to inform macroeconomic and macroprudential policymakers.

The post-2008 global financial crisis literature increasingly suggests that the relationship between FD and economic growth is nonmonotonic: FD initially fosters growth but, beyond a threshold, has a negative impact. This “vanishing effect” or “too-much-finance effect” underscores the need for further investigation. Examining interactions with institutional and regulatory environments within this assumed non-linear relationship may offer valuable insights for both policymakers­ and researchers.

Another source of inconclusiveness is the imperfection of indicators used to measure the financial system’s contribution to development. Levine (2022) notes that the literature remains inconclusive on whether finance causes growth and, if so, through what mechanisms — attributing this to limitations of FD indicators in empirical research. Many studies use the size of financial institutions or financial markets as proxies for FD. However, a more comprehensive measure is needed — one that incorporates financial institutions (FI), financial markets (FM), and their respective dimensions: access, depth, and efficiency.

Among these, financial depth is the most common proxy. Credit to the private sector is standard in cross-country regressions, but it is not comprehensive, as it fails to capture the diverse mechanisms through which the financial system supports growth (Beck, 2009). Given these limitations, the research landscape reveals a significant gap. A more nuanced measure of FD is required to assess whether a country’s financial system aligns with its macroeconomic conditions and institutional framework.

Much of the literature on FD and growth has focused on depth, neglecting access and efficiency. As a result, policymakers and regulators are often left with a one-dimensional perspective. This study addresses that gap by using a detailed FD index that includes both FI and FM, further decomposed into depth, access, and efficiency for each component. This comprehensive framework, developed by Sahay et al. (2015), provides a robust foundation for evaluating FD.

A further challenge is estimation uncertainty, which contributes to inconclusive empirical findings. Botev et al. (2019) emphasize that variation in estimation techniques­ and model specifications has led to divergent conclusions. To ensure ­robustness, this study employs the panel-corrected standard errors (PCSE) method to estimate the effects of FD in both developed and developing economies. Additionally, the generalized method of moments (GMM) technique is applied to address endogeneity concerns, thereby enhancing the credibility of the empirical results.

In line with Adam Smith’s proposition, two primary streams of thought under­pin the relationship between financial development and economic outcomes. The first stream considers FD as a supply-leading productive input that fosters economic growth. Commonly referred to as the supply-leading hypothesis, this remains the dominant perspective. Because financial markets comprise various intermediaries, it is elaborated through multiple sub-theories, including finance theory, market-based theory, bank-based theory, law-and-finance theory, and financial-services-based theory.

Renewed interest in the finance — growth nexus was sparked by the development of endogenous growth models grounded in the supply-leading hypothesis. The seminal work of Diamond and Dybvig (1983) laid the foundation for models demonstrating how financial systems influence growth by managing liquidity risk. Among the pioneers were Lucas (1988), Romer (1988), and Rebelo (1991).

Numerous studies provide empirical support for the supply-leading hypo­thesis, emphasizing the role of FD in promoting economic growth. A foundational contribution is King and Levine (1993), who found strong empirical evidence linking FD to economic expansion. Several subsequent studies reinforce these findings, including Ullah et al. (2024b) for developing countries, Afonso and Blanco-Arana (2022), Tran et al. (2021), Tang and Abosedra (2020), and Ustarz and Fanta (2021).

A significant body of research, however, identifies excessive credit expansion as a driver of economic volatility (Jorda et al., 2011; Kaminsky and Reinhart, 1999; Schularick and Taylor, 2012). Demetriades et al. (2023) observe that private credit negatively affects growth, supporting the argument that excessive financial depth can be destabilizing. Rousseau and Wachtel (2011) attribute the “vanishing effect” of FD to high financial depth, which fuels inflation and undermines banking-sector stability. Other researchers examine household credit and find it more influential for volatility than private-sector credit (Beck et al., 2014; Sassi and Gasmi, 2014; Angeles, 2015).

Recent studies also highlight that financialization affects EV through various channels. Yilmaz (2024) show that enhanced FD, particularly through improved banking and capital markets, can reduce volatility by improving risk management and capital allocation, although in some cases it may also increase exposure to systemic shocks.

Concerns about excessive financial deepening date back to Minsky (1974) and Kindleberger (1978), who suggested that unchecked financial expansion contributes to macroeconomic volatility. Deidda and Fattouh (2002, 2008) find that the positive relationship between financial depth and growth reverses during financial-system transitions, particularly when economies shift from bank-based to market-based systems.

Another mechanism through which financial deepening contributes to volatility is monetary policy. A contractionary stance can raise interest rates, limit credit access for small businesses, and increase systemic vulnerability (Beck et al., 2000). Some studies also observe that FD increases systemic risk, amplifying the potential for domestic financial crises (Rousseau and Wachtel, 2011).

The size of the financial system is also correlated with economic volatility. Beck et al. (2014) find that in high-income countries, a well-developed financial system is positively associated with volatility. Some researchers argue that firms with high collateral crowd out lower-productivity projects, ultimately slowing growth (Van Wijnbergen, 1983; Buffie, 1984). Rodrik (1998) and Stiglitz (2002) further contend that volatility in capital flows at high levels of FD undermines both economic and financial stability.

Some studies focus on financial markets as a source of volatility. Bernanke and Gertler (1990) and Bernanke and Blinder (1992) find that low-net-worth borrowers­ rely heavily on external finance, which increases agency costs and financial fragi­lity. This reliance amplifies real-sector shocks through financial‑accelerator effects during distress (Beck et al., 2000; Ibrahim and Alagidede, 2017). Acemoglu and Zilibotti (1997) suggest that investment indivisibility leads to concentrated financial risk, increasing volatility. Kiyotaki and Moore (1997) argue that capital-market imperfections exacerbate temporary productivity shocks by reducing borrowers’ net wealth, particularly among credit-constrained firms. Rajan (2005) finds that financial-innovation-led expansion contributes to tail risks, which individual investors often fail to anticipate. Abbas and Iftikhar (2016) observe that financial-sector instability increases volatility in industrial growth.

The securitization of lending portfolios has also been linked to greater instability­. Dell’Ariccia et al. (2012), Keys et al. (2010), and Mian and Sufi (2009) find that securitization weakens lending standards, resulting in higher delinquency rates. Similarly, Ashcraft and Santos (2009) and Gennaioli et al. (2012) note that financial engineering techniques, designed to match securities with risk-averse investors, can obscure critical risks and increase overall fragility. Aizenman and Pinto (2005) and Le et al. (2023) find that financial-sector expansion initially increases EV, although it may enhance growth in the short term due to a favorable risk — return trade-off. Over time, however, this volatility can reduce long-term growth by lowering investment and consumption.

Despite these concerns, some researchers argue that financial-sector development can reduce EV. This perspective is based on the premise that innovative financial structures allow greater risk-sharing, reduce financial constraints, and enhance firms’ ability to absorb shocks. These mechanisms also promote consumption smoothing, stabilizing household spending patterns. Denizer et al. (2000) find that in economies with highly developed financial systems, investment-growth volatility is significantly lower. Larrain (2006), Raddatz (2006), and Park (2015) also report that the development of financial institutions reduces output volatility in the industrial sector.

Beck et al. (2014) conclude that financial intermediation supports long-term growth while reducing volatility. Similarly, Alagidede and Ibrahim (2016) observe that financial-sector advancements dampen real-sector shocks. Liu and Yang (2016) find that financial deepening reduces macroeconomic volatility up to a threshold. Kapingura et al. (2022) also report that a well-developed financial system — where both financial institutions and markets operate effectively — contributes to lower macroeconomic volatility.

Several studies further suggest that financial deepening reduces the impact of external shocks, helping stabilize macroeconomic conditions (Iwasaki et al., 2020; Levine and Warusawitharana, 2021). Manganelli and Popov (2015) find that FD helps reduce aggregate volatility, while Xue (2020) observes that financial deepening decreases volatility in advanced economies.

Some researchers propose a U-shaped relationship between financial-sector development and EV. Alatrash et al. (2014) argue that in well-developed financial systems, FD initially reduces growth volatility, but beyond a threshold it begins to increase it. These findings are consistent with Easterly et al. (2000), Kunieda (2008), Dabla-Norris and Srivisal (2013), and Sahay et al. (2015).

Aghion et al. (2005) find that economies at an intermediate stage of development experience higher instability, while Acemoglu and Zilibotti (1997) argue that underdeveloped economies are most vulnerable due to limited investment diversification. However, another strand of research finds no significant relationship between FD and growth volatility. Beck et al. (2006) find no evidence that financial-sector development influences volatility, while Xu (2007) reports that FD does not significantly affect volatility in emerging economies. Despite these findings, the literature on this topic remains limited.

Levine (2021) emphasizes that shortcomings in FD measurement hinder definitive conclusions regarding whether and how finance drives growth. The lack of a universally accepted indicator has contributed to inconsistencies in empirical findings, making it difficult for policymakers to formulate effective strategies based on FD metrics.

A review of thirty significant studies on FD and growth reveals considerable variation in measurement approaches. As shown in Table 1, the estimates used to quantify FD differ widely. Common indicators include private credit-to-GDP, credit extended to private enterprises relative to total domestic credit, and mone­tary aggregates such as M1, M2, and M3 as percentages of GDP. Other studies employ measures such as bond-market development, private credit provided by domestic commercial banks-to-GDP, bank credit-to-GDP, and private-credit-to-deposit-money ratios. This spectrum of FD metrics underscores the lack of a standardized approach, complicating cross-country comparisons and limiting the generalizability of findings.

Beyond measurement inconsistencies, another shortcoming is the over­emphasis on financial depth as the sole indicator of FD. While depth — defined as the size of financial institutions and markets relative to GDP — is important, it does not fully capture the complexity of FD. The literature increasingly re­cognizes three core dimensions: financial depth, financial access, and financial efficiency. Financial access refers to the availability and affordability of financial services, ensuring that individuals and businesses can participate in financial systems. Financial efficiency measures how effectively institutions allocate capital, manage risk, and support economic activity.

The failure to incorporate financial access and efficiency into FD measurement represents a major gap. This omission limits policymakers’ ability to design ­holistic financial policies that promote inclusive and sustainable development. Most existing research focuses predominantly on depth, overlooking the essential contributions of access and efficiency in shaping overall effectiveness. Addressing this gap — by integrating all three dimensions (depth, access, and efficiency) into analysis — enables a more comprehensive understanding of the finance — growth relationship and supports the formulation of policies aimed at enhancing financial stability and long-term growth.

Traditional measures — such as credit-to-GDP and stock-market capitalization ratios — offer narrow proxies that do not capture the full scope of financial develop­ment. A more robust approach evaluates these aspects using composite indices with well-defined sub-indices, yielding a more accurate estimate of FD. Moreover, FD should not be confined to a single dimension (depth). A holistic measure must incorporate financial efficiency and financial access, which are equally critical for understanding how financial systems affect economic performance.

Recognizing these gaps, this study adopts a comprehensive index that evaluates both financial institutions and financial markets, incorporating depth, efficiency, and access as key dimensions through structured sub-indices.

The divergence in empirical findings further highlights the complexity of the relationship between FD and economic volatility, underscoring the need for systematic investigation. Table 2 illustrates the inconclusive nature of existing research, showing how different studies reach conflicting conclusions regarding the impact of FD on volatility. This variation suggests that the effect of FD may be context-dependent — shaped by institutional quality, regulatory frameworks, and broader macroeconomic conditions. Therefore, a more nuanced understanding of threshold effects is essential to determine whether excessive financialization is universally detrimental or whether its consequences differ across economic environments.

The literature shows that the relationship between financial development and economic volatility is both inconclusive and underexplored. Few studies investigate which specific dimensions — depth, access, or efficiency — drive heightened volatility, leaving a significant research gap. This study addresses that shortcoming. Moreover, recent research on FD and economic growth, particularly in the post-Global Financial Crisis (2008) era, has largely focused on developing countries or single-country cases (e.g., Karagol et al., 2022; Sahay et al., 2015; Bahri et al., 2018; Oro and Alagidede, 2018). This narrow scope is notable given that the crisis originated in developed countries with advanced financial markets and institutions. The lack of a comprehensive analysis of these economies may hinder effective policy design — an observation that also applies to the relationship between FD and economic-growth volatility.

Overall, the existing literature highlights a paradoxical and inconsistent relationship between financial development, economic growth, and its volatility. Empirical studies report positive, negative, U-shaped, and insignificant relationships. These inconsistencies present a dilemma for policymakers, who must decide whether to promote FD as a tool for growth or to curtail it to prevent imbalances that could destabilize the economy.

The empirical literature provides mixed evidence — both negative and positive — regarding the relationship between FD and EV. Building on this ambigui­ty, we hypothesize a nonlinear relationship between FD and EV: FD initially promotes stability by reducing EV, but beyond a threshold it may increase volatility.

H1: There is a nonlinear relationship between FD and EV in developed countries. H2: There is a nonlinear relationship between FD and EV in developing countries. (see Fig. 1).

Fig. 1.

Research flow diagram. Source: Compiled by the author.

This study examines a nonlinear relationship between financial development (FD) and economic growth volatility (EV) across two country groups — developed­ and developing — using panel data and two estimation strategies: PCSE and GMM. The main independent variable, FD, is measured with a composite index comprising three sub-dimensions — depth, access, and efficiency — each defined separately for FI and FM.

The empirical model is represented by Equation (1), which links the dependent variable, EV, to FD and controls for country j at time t:

$E{V}_{j,t}={\gamma }_0+{\gamma }_{1}F{D}_{j,t}+{\gamma }_{2}F{D}_{j,t}^2+\sum _{m=1}^n{\delta }_m{X}_{m,j,t}+{ϵ}_{j,t}$ (1)

Equation (1) tests Hypotheses 1 and 2 regarding nonlinearity between FD and EV. A positive coefficient on the quadratic term (γ₂) indicates a U-shaped relationship (stability at lower FD and higher EV beyond a threshold), whereas a negative γ₂ indicates an inverted-U. The turning point is given by −γ₁/(2γ₂) when γ₂ ≠ 0.

Diagnostic tests indicate autocorrelation, heteroskedasticity, and cross‑sectional dependence; therefore, we employ PCSE to estimate Equation (1). As a robustness check and to address potential endogeneity, we also estimate the model using GMM.

The descriptive statistics are reported in Table 4. The key independent variable­, financial development (FD), has a mean of 0.456 (std. deviation 0.202; range 0–1.097). The panel spans 60 countries (30 developed and 30 developing­) over 1981–2022, covering multiple cycles (booms, recessions, and crises). The dispersion in FD reflects the heterogeneous economic conditions in the sample.

For the linear terms of the financial-development components — financial institutions’ depth (FID), access (FIA), and efficiency (FIE), and financial markets’ depth (FMD), access (FMA), and efficiency (FME) — dispersion is moderate. Notably, FIA and FIE have standard deviations below their means, indicating relatively tight distributions across countries.

By contrast, for the squared terms of the market subindices — FMD², FMA², and FME² — the standard deviations exceed the means, suggesting that higher levels of market development are associated with greater variability. This pattern is consistent with threshold-type, volatility-enhancing effects at elevated levels of market development.

An inspection of skewness and kurtosis shows no extreme departures from symmetry or thin/thick tails: most variables have skewness within ±2 (Hair et al., 2022), and kurtosis values are moderate. As expected, squared terms (e.g., FD² components) are more leptokurtic, a common feature of transformed financial indices. These features do not undermine inference given our use of PCSE and GMM. Summary statistics for the dependent variable, EV (realized volatility of per-capita GDP growth), indicate meaningful dispersion across countries and years (Table 4). Variance inflation factors (VIFs) for linear regressors are all < 2, indicating that multicollinearity is not a concern and that predictors contribute distinct explanatory power.

The results indicate that in developed economies economic volatility rises at higher levels of financial development. A decomposition shows that financial depth in both banking and market segments is the main driver of this effect, consistent with the “too-much-finance” view: beyond a threshold, additional deepening ceases to raise productivity and instead heightens fragility.

On the banking side, abundant credit can misallocate resources toward saturated sectors, with surplus liquidity spilling into speculative, non-productive uses and inflating asset-price cycles. Concentrated banking structures may also exclude collateral-constrained entrepreneurs, dampening innovation and entrenching inequality. On the market side, excessive depth encourages complex instruments that obscure underlying risk, raising exposure to tail events.

Policy for developed economies. Financial deepening should be balanced with real-sector needs and accompanied by guardrails. Useful tools include (i) targeted credit-allocation mechanisms that tilt a portion of private credit toward innovative, high-potential firms; (ii) state-backed credit-guarantee schemes to relax collateral constraints for start-ups; and (iii) borrower-based macroprudential limits and countercyclical buffers to restrain leverage cycles. Equity-market exuberance should not dominate sectoral growth; a balanced architecture that supports both bank-based and market-based intermediation fosters resilience.

Access and efficiency at high development levels. In line with our estimates, financial-institution access (FIA) and efficiency (FIE) in developed economies display inverted U-shapes: broadening inclusion and lowering intermediation frictions are stabilizing at low-to-moderate FD, but can become destabilizing when very extensive (for example, if standards loosen or risk migrates to weaker borrower segments). Policy should therefore expand access with safeguards (e.g., proportional affordability tests, stress-tested digital onboarding) and pursue efficiency gains that do not erode underwriting quality. On the market side, widening participation (a larger free float beyond the top-ten firms) can diversify risk without encouraging excessive churn.

In developing economies, EV declines at higher FD. Development in both financial institutions and markets contributes to this stabilizing effect, plausibly because credit is more often directed to productive investment and markets are less saturated with complex, high-risk instruments. Sequencing matters: deepen access and market infrastructure while building supervisory capacity, so that later-stage risks are contained. As shown by recent evidence, improving regulatory­ quality can moderate volatility in bank-centric environments (Ullah et al., 2024a). Central banks, supervisors, and market authorities should adopt macroprudential frameworks suited to local structures (countercyclical capital buffers, sectoral risk weights, foreign-currency borrowing limits).

Interpretation. The supply-leading hypothesis on its own does not capture the nonmonotonic pattern we observe. In developed economies, higher FD is associated with greater EV at advanced stages — reflecting risk-taking, complex instruments, and expansive credit cycles. In developing economies, sustained deepening tends to stabilize EV, consistent with finance supporting productive investment and diversification.

Limitations and future research. We focus on FD as a determinant of EV. Other forces — especially financial innovation (e.g., securitization, derivatives, digital credit) — may materially shape volatility. Future work should integrate direct measures of innovation intensity, interact them with institutional quality and macroprudential regimes, and quantify threshold (turning-point) levels at which benefits give way to risks.