Digital transformation doesn't deliver equal economic gains everywhere

Digital infrastructure and financial development are both linked to stronger household consumption, but their combined effect is not as straightforward as policymakers may assume, says a new study by researchers from Imam Mohammad Ibn Saud Islamic University. Published in Economies, the study finds that digital access can support domestic economic activity, but its marginal effect weakens as financial systems become more developed.

The study, Digital Infrastructure, Financial Development, and Economic Activity: Evidence of Nonlinear Interaction Effects from G20 Countries, assessed G20 countries from 2005 to 2023 to map how internet use, financial development and their interaction shape household consumption. Internet usage and financial development were found to be positively associated with household consumption, while the interaction between the two is negative and statistically significant, pointing to diminishing gains from digital infrastructure in more financially mature economies.

Digital access supports domestic consumption

Digital payments, e-commerce, mobile connectivity and platform-based services have changed the way households buy goods, access services and participate in daily economic activity. The shift accelerated during the COVID-19 pandemic, when remote transactions and digital payments became crucial to economic life.

The researchers focus on household final consumption expenditure rather than gross domestic product because digital payments and online platforms operate most directly through transactions and domestic demand. The study treats household consumption as a better way to capture the economic effects of digital infrastructure than broader output measures alone.

The analysis uses an unbalanced panel of G20 economies, including advanced and emerging countries. It gives the model a wide range of digital, financial and economic conditions. The dataset draws on World Development Indicators and covers variables including household consumption, internet use, mobile subscriptions, secure internet servers, domestic credit to the private sector, GDP per capita, inflation and urbanization.

The results show that internet usage is positively and significantly associated with household consumption. Practically, greater internet access appears to support consumption by reducing transaction frictions, improving access to digital platforms and making it easier for households to buy goods and services.

Financial development also shows a positive link with consumption. Deeper financial systems can help households access credit, manage liquidity and smooth spending over time. It supports the long-standing view that financial systems matter for economic participation, especially when they help households move money, borrow, save and transact more efficiently.

Financial development changes digital returns

The effect of digital infrastructure depends on the financial environment around it. When internet usage and financial development are examined together, the interaction term is negative and statistically significant, which means the positive effect of digital infrastructure on household consumption weakens with rising financial development.

This doesn't mean digital infrastructure becomes unimportant in advanced financial systems. Rather, it suggests that its incremental payoff becomes smaller when financial services are already broad, efficient and widely available. In countries where financial systems are less developed, digital infrastructure can help fill gaps. Internet access and digital payment systems may reduce barriers to transactions, expand access to services and help households participate in markets that were previously harder to reach.

In more financially mature economies, the same digital expansion may add less because many transaction channels, payment systems and credit mechanisms already exist. The study describes this as a case of diminishing marginal effects. Digital infrastructure still matters, but its additional contribution is shaped by what the financial system already provides.

The authors also test the marginal effect of internet use at different levels of financial development. The estimates show stronger positive effects at lower levels of financial development, weaker effects at average levels and slightly negative marginal effects at higher levels. The pattern strengthens the view that digitalization works differently across financial environments.

The findings challenge one-size-fits-all digital transformation policies. Expanding digital infrastructure may have stronger demand-side benefits in countries where financial access and transaction efficiency remain limited. In countries with mature financial systems, policy may need to focus less on basic expansion and more on quality, security, interoperability and efficiency.

Results point to consumption, not trade openness

The study uses second-generation panel methods designed for highly connected economies. These methods account for cross-country dependence, meaning they recognize that G20 economies are affected by shared shocks such as global financial cycles, technology diffusion and worldwide shifts in digital adoption.

The researchers first test the data for cross-sectional dependence and long-run relationships. They then use the Common Correlated Effects Mean Group estimator, which allows country-specific differences while controlling for common global factors. Robustness checks include alternative specifications, trade openness as another dependent variable, and stock market capitalization as an alternative measure of financial development.

The results remain stable when the model excludes some digital infrastructure indicators and focuses on internet usage and financial development. Internet usage and financial development remain positive, while their interaction remains negative. The interaction effect also holds when stock market capitalization is used as a market-based proxy for financial development.

However, the pattern does not carry over to trade openness. When trade openness is used as the dependent variable, internet usage, financial development and their interaction are not statistically significant, suggesting that the observed relationship is mainly tied to domestic household consumption rather than external trade activity.

Digital payments and online platforms often influence how households spend inside an economy, not necessarily how much an economy trades internationally. The study thus argues that consumption-based indicators are better suited for examining the transaction effects of digital infrastructure.

The findings also show that not all digital infrastructure measures behave the same way. Internet usage has a clear positive association with household consumption, while mobile subscriptions and secure servers show negative associations in the baseline model. The authors interpret this as evidence that digital infrastructure is not a single uniform force. Different indicators may capture different stages of digital saturation, infrastructure availability or transaction intensity.

Inflation and urbanization also show negative associations in the baseline model, suggesting that macroeconomic instability and structural pressures may constrain consumption dynamics. GDP per capita is not statistically significant in the main specification, possibly because income effects overlap with digital and financial variables in the model.

Implications and limitations for digital policy

Digital infrastructure strategies should be tailored to the financial maturity of each economy, the study recommends. Policymakers cannot assume that adding more digital infrastructure will produce the same economic effect everywhere. For countries with less developed financial systems, digital infrastructure may help widen access to financial services, reduce transaction costs and strengthen household participation in digital consumption. In those settings, expanding internet access, digital payments and secure transaction systems may have a stronger impact on domestic demand.

For more financially developed economies, the priority may be different. Since basic digital and financial systems are already widely available, the gains may come from improving service quality, cybersecurity, interoperability, regulatory coordination and the efficiency of digital financial systems.

When it comes to financial inclusion policy, digital tools can support economic participation, but their effect depends on whether households can connect digital access with usable financial services. A digital platform alone may not be enough if households lack credit access, trust, digital literacy or secure payment channels.

It should be noted that the research uses aggregate country-level data, which may hide important differences between households, firms, income groups and regions. The indicators used for digital infrastructure may not fully capture newer parts of the digital economy, including fintech adoption, platform-based business models, mobile wallets and digital payment intensity.

The authors also do not fully rule out endogeneity. Household consumption may influence digital infrastructure and financial development, not only the other way around. While the study uses methods designed to handle cross-country dependence and heterogeneity, the results should be read as evidence of long-run associations rather than definitive proof of causality.

Future research could use household-level or firm-level data to examine who benefits most from digital infrastructure and how financial access shapes those benefits. More detailed measures of digital payments, fintech use, platform adoption and regulatory quality could also sharpen the analysis. Additional work could examine whether threshold effects exist, where digital infrastructure produces strong gains up to a certain level of financial development before the returns begin to flatten.