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

The shadow economy's determinants and interactions with economic growth, financial development, inequality, corruption, and trade liberalization are widely studied. While offering employment, its expansion is detrimental, worsening inequality, hindering growth, and aggravating tax evasion. This paper defines the shadow economy as economic activities concealed from regulatory authorities for monetary, regulatory, and institutional reasons. Institutionalized democracy is defined as the establishment of democratic practices and organizations. The research question is whether improved institutionalized democracy reduces the shadow economy in Uganda. Uganda provides a suitable case study due to its turbulent political history, strides in governance improvement, and the lack of research on this relationship in the African context. This study contributes by extending research on direct democracy and the shadow economy to institutionalized democracy; examining the policy implications of this relationship; investigating this relationship in an African context; and focusing on the political system's impact on the shadow economy, a gap in existing literature. While building on Teobaldelli and Schneider (2013), this study differs by using a different measure of democracy, focusing on Uganda, and employing the ARDL bounds testing approach to cointegration.

A substantial portion of economic activity occurs in both formal and informal sectors globally. Estimates show a large and persistent shadow economy across developed and developing countries, contradicting earlier suggestions that informality is a feature of underdevelopment. The persistence is attributed to the formal sector's inability to generate sufficient jobs and the additional costs associated with formalization (increased taxation and burdensome regulation). While some argue that the shadow economy complements the formal economy, others highlight its negative impacts, such as increased income inequality, reduced tax revenue, distorted resource allocation, and biased macroeconomic measures. While previous research has explored financial and economic dimensions, this study focuses on the political and governance aspects of the shadow economy.

This study uses annual time series data from Uganda (1991-2015). The dependent variable is the shadow economy (se15), measured as a percentage of GDP (Medina and Schneider, 2018). The main explanatory variable is the institutionalized democracy index (demo) from the Center for Systemic Peace. Control variables include political fractionalization (frac), GDP growth (gw), government spending as a share of GDP (gov/gdp), domestic credit to the private sector (dob), and regime durability (dur). Descriptive statistics show a negative correlation between the shadow economy and institutionalized democracy. The ARDL bounds testing approach to cointegration is used to examine the long- and short-run relationships. This method is robust, applicable regardless of sample size, and can correct for endogeneity. The ARDL model is specified, and stationarity tests are conducted using ADF and PP tests. The ARDL bounds test assesses the long-run relationship, followed by estimation of short-run and long-run coefficients. The optimal lag length is determined using the Schwartz-Bayesian criterion (SBC). Residual diagnostic tests (Breusch-Godfrey Serial Correlation LM Test, Heteroskedasticity Test, and Normality test) are conducted to ensure reliable results. For robustness, FMOLS and DOLS methods are also used.

Stationarity tests reveal that variables are stationary after first differencing. The ARDL bounds test indicates a cointegration relationship between the variables. Long-run results (ARDL, FMOLS, and DOLS) show a negative and statistically significant relationship between institutionalized democracy and the shadow economy. A one-unit increase in institutionalized democracy reduces the shadow economy by 3.399 units (1% significance level). Economic growth negatively affects the shadow economy (10% significance level), while government expenditure positively affects it (1% significance level). Financial development negatively affects the shadow economy (1% significance level), and regime durability negatively affects it (1% significance level). Fractionalization is negatively correlated with the shadow economy. Short-run results show a similar negative relationship between institutionalized democracy and the shadow economy, with a one-unit increase in democracy reducing the shadow economy by 3.663 units (1% significance level). Financial development (1% significance level), regime durability (1% significance level), and fractionalization (1% significance level) are all significantly negatively correlated with the shadow economy in the short run. Government expenditure is positively correlated with the shadow economy in the short run (1% significance level). The error correction term is negative and statistically significant (1% level), indicating a 94.07% adjustment speed to long-run equilibrium. Stability tests (CUSUM and CUSUMQ) confirm the stability of the ARDL model's coefficients.

The findings support the hypothesis that improved institutionalized democracy reduces the shadow economy. The negative correlation in both the short and long run suggests that stronger democratic institutions lead to better resource allocation, improved citizen welfare, and reduced incentives for informal activities. The impact of economic growth, financial development, and regime durability aligns with previous studies. The positive impact of government expenditure on the shadow economy indicates that high taxation can drive businesses underground. The significant and negative error correction term suggests a relatively fast adjustment towards long-run equilibrium. The robustness checks using alternative econometric methods support the findings.

This paper demonstrates a robust negative relationship between institutionalized democracy and the shadow economy in Uganda. Policy implications suggest focusing on governance reforms to enhance resource allocation and shifting from addressing immediate causes of informality to implementing democratic reforms. Future research could explore interactions between other variables and the shadow economy, investigate causes and consequences of the shadow economy in Africa, develop theoretical linkages, and analyze shadow economy activity in specific sectors.

The study focuses solely on Uganda, limiting the generalizability of findings. The use of a single measure for institutionalized democracy may not capture all relevant aspects. The time series nature of the data could be subject to omitted variable bias. Further research with a broader scope and more comprehensive datasets is needed.