Kalilla Abdullayev

Type of the article: Research Article

AI is widely regarded by the IMF and the World Bank as a catalyst for growth. AI should be understood as a multidimensional socio-technical system embedded across institutions, industries, and society. Its economic contribution depends on which pillars of the national AI system expand (e.g., R&D capacity, infrastructure, governance, or social acceptance). For this reason, the seven pillars of AI development are measured by the AI Vibrancy subindices, which help avoid reliance on a single composite indicator that may conceal offsetting effects. This study examines how different pillars of the national AI ecosystem shape the architecture of the knowledge economy and its economic outcomes by estimating heterogeneous within-country associations between GDP per capita and seven AI ecosystem pillars, operationalized through AI Vibrancy subindices, using a balanced panel of 36 countries with complete data over the period 2020–2023. Fixed- and random-effects models are estimated using heteroskedasticity-robust and Driscoll-Kraay standard errors. The results indicate that, within countries over time, the R&D (β = –5.676, p < 0.001) and Infrastructure (β = –16.306, p < 0.001) subindices have strong and statistically significant negative associations with GDP per capita, while Public Opinion shows an adverse effect that is significant at the 5% level under heteroskedasticity-robust inference (β = –9.126, p = 0.040) and marginally significant under Driscoll-Kraay inference (p = 0.054). Responsible AI exhibits a marginally positive association (β = 5.773, p = 0.065) in the Driscoll-Kraay specification, whereas Economy, Education, and Policy & Government show no significant within-country effects.

Type of the article: Research Article

Abstract
Innovative marketing increasingly requires reliable market intelligence signals that reduce uncertainty, support product positioning, and guide venture financing in technology-intensive green and digital markets. This study aims to assess how patent-based technological signals shape innovative market development by predicting the formation and venture financing of green and digital energy start-ups, while also examining whether entrepreneurial market entry and funding stimulate subsequent patenting activity. The empirical analysis is based on a balanced panel of 146 countries for 2000–2023, combining IEA energy start-up and funding indicators with OECD patent data. The empirical strategy follows a sequential design: descriptive statistics and log1p transformations are used to characterize the data; Dumitrescu–Hurlin panel Granger causality tests provide the main evidence on predictive causality; and PVAR, multiple-testing corrections, PPML and TWFE models are used as complementary robustness and dynamic checks. The results show highly concentrated innovative market development: average green and digital energy start-up activity is around 7 per country-year, while the median is 0 for both indicators. The Dumitrescu–Hurlin tests reveal 69 significant relationships out of 120, with stronger evidence for patents predicting start-up formation and funding than for the reverse direction. These findings remain robust after Benjamini–Hochberg correction and after excluding numerically extreme statistics. TWFE results support the positive association between climate adaptation and ICT-mitigation patents, digital energy start-up formation and early-stage digital funding, while PVAR models provide only complementary dynamic evidence and are interpreted cautiously due to stability limitations in the main GMM specification. Country fixed effects indicate that Ukraine has a more favorable estimated structural position for digital energy start-up formation than Kazakhstan and Armenia.

Type of the article: Research Article

Abstract
Weak risk governance and legacy non-performing loans have repeatedly destabilized banks in emerging markets, making the payoff from regulatory strengthening of internal control and internal audit a first-order policy question. This study aims to determine whether Kazakhstan’s 2019 risk-management and internal control requirements (National Bank Resolution No. 188), whose core assurance mechanism is a strengthened internal audit function, improved the financial stability of the country’s commercial (second-tier) banks, and to identify the channel of this effect. The analysis applies a continuous-treatment difference-in-differences design to an annual panel of 37 banks over 2015–2025 (287 bank-years), interacting each bank’s pre-intervention (2016–2019) weakness with the post-intervention period and measuring stability by the log Z-score. The results show that the stabilizing effect is concentrated in the asset-quality channel: banks that entered the post-2020 period with a one-standard-deviation greater non-performing-loan weakness recorded a 0.38-0.41 log-point (roughly 40-50%) larger post-intervention increase in the Z-score (p < 0.01), whereas composite pre-intervention weakness yields no robust effect under any weighting scheme. The effect emerges with a multi-year lag, becoming statistically detectable only toward the end of the sample period, and operates through capital rebuilding. Because pre-intervention weakness strongly predicts market exit (Cox hazard ratio ≈ 3.95), the estimates are a lower bound on the intervention’s full association with sector stability. The findings imply that supervisors should target credit-risk governance, loan classification, timely recognition of problem loans, and provisioning, where strengthened internal control and audit yield the largest stability gains.

Type of the article: Research Article

Abstract
Digital financial infrastructure serves as a key structural correlate of small and medium enterprise (SME) credit availability and systemic financial development across post-Soviet transition economies. This study evaluates the empirical associations between technological connectivity, institutional policy readiness – specifically Digital ID and ISO 20022 standardization – and SME credit access (CreditLine) as the primary outcome of interest, alongside auxiliary supporting analyses of population-level financial inclusion (Ic_Y_norm) and macroeconomic output (lnGDPperCapita). Utilizing a panel dataset for 14 post-Soviet countries from 2011 to 2023 (N = 182) sourced from the World Bank WDI, Global Findex, and official central bank reports, the empirical framework relies on One-Way Country Fixed Effects (FE) specifications with Driscoll–Kraay robust standard errors to account for spatial and temporal dependence, supplemented by a Two-Way Fixed Effects (TWFE) sensitivity analysis. Mobile connectivity demonstrates a positive baseline association with SME credit availability, while internet penetration is identified as a primary empirical correlate of population-level financial inclusion across specifications. The protection of legal rights correlates with enhanced SME credit access, whereas strict AML/CFT compliance shows a strong negative association. Institutional messaging standardization via ISO 20022 exhibits heterogeneous structural differential associations: while positively correlating with SME credit access, it reflects a statistically significant negative association with broader population inclusion. Robustness estimations indicate that exogenous regional shocks (COVID-19, geopolitical conflicts, and energy crises) and macroeconomic slack (inflation, unemployment) correspond to systemic contractions in credit availability.