Institutionalized democracy and the shadow economy in the short- and long-run: empirical analysis from Uganda

The study examines whether improvements in institutionalized democracy reduce the size of the shadow economy in Uganda. The shadow economy—defined as all economic activities concealed from regulatory authorities for monetary, regulatory, or institutional reasons—has drawn significant attention due to its implications for growth, inequality, and tax evasion. Institutionalized democracy is understood as the process by which democratic practices become well established and widely accepted. Using annual data for Uganda from 1991–2015 and the ARDL bounds testing approach, the paper investigates both short- and long-run relationships between democracy and informality. Uganda provides a pertinent context given its turbulent political history, gradual democratization in recent decades, and documented prevalence of informality. The research addresses a gap in the literature by focusing on political and governance dimensions—specifically institutionalized democracy—in an African context, and asks whether strengthening democratic institutions mitigates informality.

The literature indicates that informality remains substantial and persistent worldwide, averaging about 31% of GDP across 157 economies (Medina and Schneider, 2019). Regional variation shows high levels in Latin America, parts of Europe, and Africa; Uganda’s shadow economy is around 31% of GDP. Persistence of informality is linked to insufficient formal job creation, high taxation, and burdensome regulation, which raise the costs of formality and encourage underground activity. While the informal sector can provide livelihoods and complement formal sector activity, numerous studies document its downsides: increased income inequality, reduced economic growth, tax evasion, distortions in resource allocation, and biases in official macroeconomic statistics that complicate policy design. Much of the literature focuses on economic and financial determinants (e.g., growth, taxation, financial development, trade openness, corruption). Fewer studies examine how governance and political institutions shape the size of the shadow economy. This paper fills that gap by focusing on institutionalized democracy and its potential to reduce informality through better resource allocation, accountability, and institutional functioning.

Data: Annual time series for Uganda, 1991–2015. Dependent variable: size of the shadow economy as a percent of GDP (se15) from Medina and Schneider (2018). Main explanatory variable: institutionalized democracy index (demo) from the Center for Systemic Peace (range −10 to +10). Controls: political fractionalization (frac), GDP growth (gw), government expenditure as percent of GDP (gov/gdp), domestic credit to private sector by banks as percent of GDP (dob; proxy for financial development), and regime durability (dur). Data sources include World Bank World Development Indicators, World Bank’s Database of Political Institutions, and the Center for Systemic Peace.

Theoretical framework: Building on Cukierman et al. (1992) and Elbahnasawy et al. (2016), the model posits that inefficiency in tax collection (affected by political polarization and instability) increases incentives for firms to operate in the shadow economy. In politically stable and less polarized systems, stronger democratic institutions are expected to reduce tax collection inefficiencies and thus reduce informality.

Econometric strategy: Stationarity is assessed using Augmented Dickey-Fuller (ADF) and Phillips-Perron (PP) tests. The autoregressive distributed lag (ARDL) bounds testing approach (Pesaran and Shin, 1999; Pesaran et al., 2001) is used to test for cointegration and estimate long- and short-run relationships. Lag length is selected via Schwartz-Bayesian/Schwarz Information Criterion. The estimated ARDL specification includes shadow economy, democracy, and the controls. Upon finding cointegration, an error-correction model (ECM) is estimated to capture short-run dynamics and speed of adjustment. Robustness checks use Fully Modified OLS (FMOLS) and Dynamic OLS (DOLS). Diagnostic tests include Breusch-Godfrey serial correlation LM, Breusch-Pagan-Godfrey heteroskedasticity, normality, and stability via CUSUM/CUSUMQ.

Model: se15 = F(demo, frac, gw, gov/gdp, dob, dur). Optimal ARDL model selected: ARDL(1,0,0,0,0,0,0). Cointegration confirmed via bounds test. ECM includes lagged error-correction term to measure speed of adjustment.

• Cointegration: ARDL bounds F-statistic = 7.848 (T=24), exceeding upper critical bounds, indicating a long-run relationship. Residual diagnostics support model adequacy.
• Long-run (ARDL; corroborated by FMOLS and DOLS): • Institutionalized democracy (demo): −3.399 (t=−4.357, p<0.01) — improvements in democracy reduce the shadow economy. • Political fractionalization (frac): −1.418 (t=−3.594, p<0.01) — higher fractionalization associated with smaller shadow economy. • GDP growth (gw): −0.121 (t=−2.002, p≈0.063) — higher growth modestly reduces informality. • Government expenditure (gov/gdp): +0.379 (t=3.678, p<0.01) — higher fiscal burden increases informality. • Financial development (dob): −0.384 (t=−6.618, p<0.01) — more credit to private sector reduces informality. • Regime durability (dur): −0.194 (t=−3.967, p<0.01) — greater durability reduces informality. • Robustness: FMOLS and DOLS yield qualitatively and quantitatively similar coefficients and significance.
• Short-run (ECM specification): • Institutionalized democracy: −3.663 (t=−7.268, p<0.01). • Fractionalization: −1.938 (t=−4.215, p<0.01). • GDP growth: −0.096 (t=−1.741, p=0.104), not statistically significant. • Government expenditure: +0.342 (t=5.962, p<0.01). • Financial development: −0.374 (t=−10.444, p<0.01). • Regime durability: −0.181 (t=−5.483, p<0.01). • Error-correction term (ECT−1): −0.941 (t=−4.615, p<0.01), implying about 94% speed of adjustment towards long-run equilibrium.
• Diagnostics and stability: Serial correlation, heteroskedasticity, and normality tests indicate no issues; CUSUM/CUSUMQ indicate stable coefficients.

The results directly answer the research question: improvements in institutionalized democracy are associated with a significantly smaller shadow economy in both the short- and long-run in Uganda. Stronger democratic institutions likely enhance accountability, constrain executive power, and improve provision of public goods, thereby improving welfare and reducing incentives to operate informally. Democracy can facilitate efficient resource allocation, reduce tax collection inefficiencies, and foster trust and compliance, which curtail shadow activities. Complementary findings show that economic growth and financial development also reduce informality, while higher government expenditure (as a fiscal burden proxy) increases it. Political factors matter: higher regime durability and political fractionalization (as measured here) are associated with lower informality, consistent with the idea that stable, institutionalized political environments reduce uncertainty and support formal sector activity. Robustness across ARDL, FMOLS, and DOLS and supportive diagnostics reinforce the credibility of these relationships.

This paper shows that institutionalized democracy significantly reduces the size of the shadow economy in Uganda in both the short and long run. Using annual data (1991–2015) and ARDL bounds testing, corroborated by FMOLS and DOLS, the study finds strong negative effects of democracy and additional reductions from economic growth, financial development, and regime durability, whereas higher government expenditure increases informality. Policy implications include prioritizing democratic reforms that strengthen institutions and improve accountability, as well as advancing financial development and growth-friendly policies, to reduce informality. Future research directions proposed include exploring interactions between the shadow economy and other determinants, investigating causes and consequences in broader African contexts, deepening theoretical linkages, and, where data permit, sectoral analyses (e.g., agriculture, manufacturing, services) to uncover heterogeneity within the shadow economy.