Digital Infrastructure Integration and Capital-Market Resilience in a Frontier Exchange: A Longitudinal Panel and Coefficient-Decomposition Analysis of the Nairobi Securities Exchange (Cureus Journal of Business and Economics.)

Summary

• The study analyzes a balanced panel of 53 ordinary equity securities continuously listed and traded on the Nairobi Securities Exchange over the 2021-2025 period.p.5
• Sub-Saharan Africa represents the global center of mobile money, processing approximately USD 1.4 trillion in transactions in 2025.p.3
• During the macroeconomic stress period of 2023, the Kenyan shilling experienced a year-on-year depreciation of 20.4% against the US dollar under monetary tightening.p.2
• Firms in the high digital exposure tier recorded lower daily price volatility of 2.02% and higher average returns of 14.73% compared to traditional sector counterparts.p.9
• A Gelbach decomposition attributes −0.120 of the −0.099 total raw DIE coefficient gap directly to between-firm market liquidity.p.1
• Descriptive statistics across 212 firm-year observations show an average annual logarithmic equity return of 10.22% and a mean daily price volatility of 2.98%.p.8

Key insights

1. Market liquidity, rather than direct digital exposure, carries the primary association with equity price volatility during market shocks.p.1
2. A Hausman specification test strongly rejects random effects in favor of fixed effects for time-varying regressors with a test statistic of χ²(3) = 41.11 (p < 0.001).p.7
3. High digital exposure firms exhibited lower price volatility of 2.02% compared to 3.42% for moderate exposure and 3.01% for low exposure firms.p.9
4. In the correlated random effects model, the direct effect of digital infrastructure exposure on price volatility is statistically insignificant at 0.097 when controlling for firm-mean liquidity.p.10
5. Gelbach coefficient decomposition confirms that firm-mean log trading volume contributes −0.120 of the −0.099 raw coefficient change across models.p.14
6. Hypothesized moderation where digital exposure steepens the volume-return relationship was not supported, as interaction terms remained statistically insignificant across all specifications.p.12
7. Re-estimating the volatility model on an expanded unbalanced sample of 58 firms confirms the market liquidity channel with a firm-mean log volume coefficient of −0.448 (p < 0.01).p.18