Abstract
This chapter studies the key channels through which improvements in governance and anticorruption measures can boost macroeconomic performance, particularly long-run economic growth.2 It estimates the relationship between economic growth and both corruption and aggregate indicators of governance, the latter reflecting several dimensions of institutional quality, including not only a measure of corruption, but also voice and accountability, effectiveness of government policies, regulatory framework, and rule of law. It also discusses the extent to which improvements in these areas can benefit governance, including institutions responsible for fiscal policy and monetary and financial management. The empirical and policy discussions focus on developing countries, with a special emphasis on sub-Saharan Africa. The empirical findings suggest that, on average, sub-Saharan African countries lag other regions in terms of governance and perceptions of corruption. The region’s dividend from governance and anticorruption reforms could therefore be particularly high. Historical experience suggests that improving governance and reducing corruption may take considerable time and work, but the large potential payoffs would justify the effort.
Introduction
Improving governance and fighting corruption remain critical ingredients in boosting growth and development in sub-Saharan Africa. The African Union chose “Winning the Fight against Corruption” as its theme for 2018. Many incoming leaders in countries in the region routinely place good governance at the top of their agenda. Historical data suggest that sub-Saharan African countries generally lag those in most other regions in terms of corruption perceptions and governance. Sub-Saharan Africa’s average scores on corruption and governance are similar to those for the Middle East, North Africa, Afghanistan, and Pakistan region (MENAAP) but lower than for other regions (Figures 2.1.1 and 2.1.2). Thirty-six out of 45 sub-Saharan African countries score below the global average in Transparency International’s Corruption Perceptions Index (CPI) and only two of the 30 sub-Saharan Africa countries included in the International Country Risk Guide’s (ICRG) governance indicator have above-average scores (Figure 2.1.3). There is, however, significant intraregional variation in scores across sub-Saharan Africa (Figure 2.1.4).3
Corruption Perceptions and Governance in Sub-Saharan Africa and the World (2019–20)
Sources: Transparency International; ICRG; and authors’ calculations.Note: All indicators are normalized 0–100, lower (higher) score better for anti-corruption (governance). Confidence intervals are estimated around 2020 scores using standard deviations for the five-year period 2016–20.Corruption Perceptions and Governance in Sub-Saharan Africa and the World (2019–20)
Sources: Transparency International; ICRG; and authors’ calculations.Note: All indicators are normalized 0–100, lower (higher) score better for anti-corruption (governance). Confidence intervals are estimated around 2020 scores using standard deviations for the five-year period 2016–20.Corruption Perceptions and Governance in Sub-Saharan Africa and the World (2019–20)
Sources: Transparency International; ICRG; and authors’ calculations.Note: All indicators are normalized 0–100, lower (higher) score better for anti-corruption (governance). Confidence intervals are estimated around 2020 scores using standard deviations for the five-year period 2016–20.
Sample Distribution of Governance and Corruption Perceptions
Sample Distribution of Governance and Corruption Perceptions
Sample Distribution of Governance and Corruption Perceptions
Overall Effect of Governance on Growth across Regions
Source: Hammadi et al. (2019).Notes: The charts show β1+β2 (red diamonds) and 95-percent confidence intervals.***, **, * denotes significant at 1, 5, and 10 percent, respectively. LAC = Latin American and Caribbean; MENAAP = Middle East, North African and Afghanistan and Pakistan; Non-SSA = Developing and emerging non sub-Saharan African countries.Overall Effect of Governance on Growth across Regions
Source: Hammadi et al. (2019).Notes: The charts show β1+β2 (red diamonds) and 95-percent confidence intervals.***, **, * denotes significant at 1, 5, and 10 percent, respectively. LAC = Latin American and Caribbean; MENAAP = Middle East, North African and Afghanistan and Pakistan; Non-SSA = Developing and emerging non sub-Saharan African countries.Overall Effect of Governance on Growth across Regions
Source: Hammadi et al. (2019).Notes: The charts show β1+β2 (red diamonds) and 95-percent confidence intervals.***, **, * denotes significant at 1, 5, and 10 percent, respectively. LAC = Latin American and Caribbean; MENAAP = Middle East, North African and Afghanistan and Pakistan; Non-SSA = Developing and emerging non sub-Saharan African countries.This chapter investigates the relationship between economic performance and governance/corruption, with a special focus on sub-Saharan Africa. The main goal of this investigation is to test whether weak governance and corruption have any impact on economic growth. The chapter posits that corruption and weak governance undermine growth, thus acting more like “sand in the wheels”4 than “oil in the engine.”5 The chapter also examines the “grease in the wheels” view— which argues that some amount of corruption could increase bureaucratic efficiency and improve growth by mitigating red tape—by testing a nonlinear relationship between corruption and growth. Of course, low corruption and good governance are not the sole drivers of growth. There are various examples of poorly governed countries that have had episodes of strong growth driven by other factors (for example, natural resources), while others have not necessarily experienced strong growth despite their good governance.
The rest of the chapter is organized as follows: The first section, “Measurement, Stylized Facts, and Channels for a Governance Dividend,” discusses the indicators used as proxies to gauge the effectiveness of governance and some of the challenges associated with these measures, as well as the main channels through which governance can impact macroeconomic outcomes. The following section presents the empirical approach, discusses the baseline findings, and shows the various robustness checks. Finally, “Policy Implications” discusses some of the policies that could support efforts to improve governance and political economy considerations in implementing such policies.
Measurement, Stylized Facts, and Channels for a Governance Dividend
Measuring Governance
Governance is a multifaceted issue that cuts across politics, economics, and institutions. The indicators with the most significant economic ramifications include corruption (abuse of public office for private gain), government effectiveness (quality of public policies and services), regulatory quality (ability of the government to formulate and implement business-friendly policies and regulations), the rule of law (respect for contract enforcement, property rights, and law enforcement), and voice and accountability (the extent to which citizens participate in selecting their government, as well as freedom of expression, freedom of association, and a free media).6
Bringing the various governance dimensions into one indicator can be challenging because aggregating subjective measures may not fully capture the reality on the ground—since distinct attributes of governance are lumped together in one indicator. While corruption perceptions tend to be the main component of interest, many measurements of governance are broad enough to be useful proxies for determining the quality of political institutions, government regulations, and policies.
Stylized Facts
Weaker governance and higher corruption are associated with the following features:
Lower levels of development: The data show an unconditional positive (negative) correlation between governance (corruption perceptions) and real GDP per capita both in sub-Saharan Africa and the rest of the world. Similar correlations have also been found or implied in an extensive body of literature, including seminal work by Mauro (1995) and meta-analysis by Ugur and Dasgupta (2011).
Lower growth in sub-Saharan Africa: The data show a positive (negative) unconditional correlation between governance (corruption perceptions) and growth in sub-Saharan Africa.7 These correlations look weaker for the rest of the sample.
Worsened fiscal performance, as seen in expenditure levels and composition: Higher corruption perceptions are negatively associated with education spending and quality of public investment in the overall sample and more so in sub-Saharan Africa.8
Higher inflation and increased risks of financial instability: These can arise from monetary financing pressures on a central bank lacking independence, or the allocation of loans based on political connections.9
Sub-Saharan Africa lags peers in most granular measures (that is, channels) of governance. Sub-Saharan African countries have low scores across most dimensions of governance (Figure 2.2), including voice and accountability, government effectiveness, regulatory quality, and rule of law, as well as business environment.
These channels are also strongly correlated with corruption perceptions. Thus, improving governance channels could reduce corruption and therefore mitigate the costs of corruption.10 Overall, this evidence for sub-Saharan Africa suggests that governance and corruption problems are both acute and endemic in the region. Weak institutions and resource intensity could be linked to the higher incidence of weak governance and corruption in sub-Saharan Africa. The large economic rents generated by resource-rich sectors such as oil—often controlled by state-owned companies subject to political interference—can expose resource-rich countries to higher levels of corruption, especially when institutions are weak (OECD 2014; 2016).
Channels for a Governance Dividend
Sub-Saharan Africa’s governance scores do not necessarily reflect a lack of legislative and institutional frameworks, as many countries have adopted legislation that criminalizes corruption and related offences, improved their AML/CFT frameworks, and established specialized anticorruption agencies. Instead, they likely reflect limited institutional capacities, weak enforcement of existing frameworks, and a perverse interaction between weak institutions and resource intensity. Hence, in many cases, adhering to well-designed rules and regulations can represent a major step in the right direction.
Improved policies that strengthen governance through multiple channels can enhance economic performance and support social inclusion indirectly by positively affecting revenue, investment, and the financial sector. These positive impacts, and the policies that enable them, include:
Enhanced revenue mobilization through improved tax compliance:11 Customs and revenue authorities are better able to fight smuggling and illicit flows when tax officials adhere to strong governance principles. Citizens are more likely to pay their taxes when they trust the effectiveness of government spending.
More efficient government spending due to stronger budgetary processes:12 Good governance reduces the risks of deleterious shifts in the composition and quality of government spending towards items that allow for greater graft opportunities (for instance, “white elephants”). Strict adherence to public procurement and public financial management laws reduces the risk of expensively tailored contracts, thus lowering overall spending inefficiencies and fiscal risks.
Improved investment composition and business environment:13 Addressing weakness in investment procedures improves public investment management and efficiency. Removing red tape and regulations reduces the opportunities for corruption and the cost of doing business, thus supporting private investment. Streamlined regulations also reduce policy uncertainty and the risk of state capture, whereby agents pay for regulations to be tailored to their business needs. Leveling the playing field is generally conducive to firm entry, competition, and innovation.
Better central banking governance and reduced price and financial stability risks:14 Improved fiscal governance translates into reduced fiscal pressures and supports price stability by reducing the need for central bank financing or a raid on the central bank coffers. Strengthened financial supervision supports financial stability by reducing the likelihood of lending irregularities and crony capitalism.
Improved developmental outcomes and social inclusion:15 Improved spending composition and investment are likely to benefit the poor disproportionately, as they rely more on social services. Hence, improved spending on education and health can better support economic and social inclusion and reduce social vulnerabilities, as well as mitigate income inequality and poverty.
Empirical Approach and Findings
A standard growth model, augmented for measures of governance or corruption, is estimated in order to assess the impact on GDP per capita growth for 190 countries, using five-year observations over the period 1984–2015 in a system of generalized method of moments (SGMM) model. Box 2.1 briefly describes the methodology, and Annex 2.1 summarizes the data and data sources. In the baseline regressions, the following variables of interest are used: (1) two measures of governance—the ICRG’s Political Risk Rating and an aggregate measure based on Kaufmann and Kraay’s Worldwide Governance Indicator (WGI) (Kaufmann, Kraay, and Mastruzzi 2010); and (2) two indicators of corruption perceptions—Kaufmann and Kraay’s Control of Corruption Indicator (CCI) and Transparency International’s CPI. The discussion that follows focuses on sub-Saharan Africa, but it also explores the correlation between governance/corruption and growth in other regions and country groups; it further considers whether the correlation is stronger in sub-Saharan Africa compared to other regions.
All estimated growth dividends presented in this chapter are long-term gains. Measuring the exact time that is required for these gains to materialize is beyond the scope of this chapter, but it will depend to a large extent on the starting point of the countries and the overall political commitment to the process. Giavazzi and Tabellini (2005) estimate that economic and political liberalization, which is typically associated with improved governance, takes at least three years to affect governance and corruption perceptions.
Estimating the Impact of Corruption and Governance on GDP per Capita Growth
A standard growth regression is estimated, augmented for governance or corruption and using an unbalanced panel comprising of 190 countries over the period 1984–2015. The baseline specification is the following:
where g is real GDP per capita growth, GOV is governance or corruption perceptions, SSA is a dummy variable for sub-Saharan Africa, X is a column vector of country-specific explanatory variables that are assumed to be strictly exogenous in the baseline specification, τ denotes time-fixed effects, and μ and v, denote unobserved country fixed effect and error term, respectively. Subscripts i = 1, 2, …, N, and t = 1, 2, …, T index country and time, respectively. The baseline variables included in vector X have been typically considered by similar studies in the literature: initial GDP per capita (in log); physical investment (measured by gross capital formation as percent of GDP); level of education (per capita years of secondary and tertiary schooling); inflation (dummy variable if inflation is larger than 15 percent and zero otherwise); and terms of trade (percent change). See full data description in Annex 2.1. The robustness of the model has also been tested against alternative choices of controls (see following discussion).
The baseline model was estimated using a robust two-step SGMM model (Blundell and Bond 1998; Roodman 2009). The system includes the level and difference versions of specification (1). Investment and lagged growth (included in alternative specifications) have been treated as endogenous and the remainder (initial income, education, inflation, terms of trade, governance/corruption, and time and regional dummies) have been considered as exogenous in the baseline specification. For the differenced equation, the estimator uses as instruments lagged values of endogenous and predetermined variables and current and lagged values of differenced exogenous variables. For the level equation, it uses as instruments lagged values of differenced endogenous and predetermined variables and current and lagged values of exogenous variables. The absence of second-order serial correlation of the error term and validity of instrument (Hansen test of overidentifying restrictions) have also been tested for. The coefficients of interest are β1 and β2. Both are expected to be positive for governance and negative for corruption perceptions. The stylized facts set forth in the preceding discussion suggest that β1 + β2 > β1 (in absolute terms) for sub-Saharan Africa; that is, governance or corruption perceptions could potentially have a stronger marginal effect on growth in the sub-Saharan Africa region than in other regions. The interaction term would appear to encapsulate this feature. Interaction terms, including cate-gorial variables, have also been used in the literature in a somewhat similar setup to that used in this chapter.1
Because weak governance and high corruption distort social and capital spending and worsen government policies overall, the covariates gross capital formation, education, and inflation already capture some of the indirect impact of governance and corruption on growth. Therefore, β1 and β2 would measure only the direct effect on growth, after controlling for human and physical capital and the quality of macroeconomic policies.
The baseline model survives several robustness tests, including: the use of alternative estimation methods (for example, difference GMM, OLS); inclusion of additional controls (for example, a 20 percent threshold for inflation dummy, resource intensity, agriculture intensity, openness, and government consumption); and choice of the averaging window (for example, one year and three years). Furthermore, several tests were performed for potential endogeneity problems in the relationship between governance/corruption and growth. The baseline results pass most of these tests, including lagging governance and using several alternative instrumental variables such as ethnic fractionalization, settler mortality, years since independence, and a country’s latitude.2
A nonlinear relationship is also estimated between corruption and growth, which allows for testing of the “grease in the wheels” hypothesis. The finding here is that corruption seems to be more harmful to growth in sub-Saharan African countries where corruption perceptions exceed 60 in the normalized scale used in this analysis, that is, about three-quarters of all sub-Saharan country-years in the sample. The impact of corruption perceptions on growth seems less harmful if corruption problems are perceived to be small to moderate.3
1 See, for example, Gyimah-Brempong and de Camacho (2006). 2 See Hammadi and others (2019). 3 See, for example, Saha, Mallik, and Vortelinos (2017).The main findings point to an adverse correlation between weak governance/ corruption and growth that is both statistically and economically meaningful and has a higher impact in sub-Saharan Africa. These estimates are generally robust to various sensitivity checks, including to alternative measures of governance, sample period, country groupings, and specifications that control for omitted variables bias and mitigate potential endogeneity problems.
While the estimated correlations between governance/corruption and growth do not prove causality, they do suggest the following:
The impact of weak governance on GDP per capita growth is stronger in sub-Sa-haran Africa relative to the rest of the world. Based on the baseline estimates (Table 2.1), this study suggests that improving sub-Saharan Africa’s governance by about one-standard deviation16—which for an average sub-Saharan African country would result in governance converging to the world average—is associated with an increase in GDP per capita growth of about 1–2 percentage points, depending on the governance indicator, with WGI (or ICRG) at the low (or the high) end of this range. This impact is two to three times larger than for the average country in the rest of the world and stronger than that of other regions like Latin America and the Caribbean (LAC) and MENAAP, which are also perceived to have weak governance.
The impact of weak governance on GDP per capita growth in sub-Saharan Africa is comparable to or stronger than in other regions that are also perceived to face acute governance problems. The sample was split in two distinct ways: first, the baseline model was run using the overall sample but including one region at a time (alternative 1); then we use only the samples for each region (alternative 2). Based on the estimates from these specifications (Figure 2.3), the finding is that improving governance by about one standard deviation in the sub-Saharan African sample is associated with an increase in GDP per capita growth of about a similar order of magnitude as in the baseline. As in the baseline, the estimated growth payoffs are stronger for ICRG compared to WGI. This is slightly larger than the impact for other regions such as LAC and MENAAP that are also perceived to suffer from acute governance and corruption problems. Overall, these findings suggest that weak governance appears to hinder growth in other regions, but apparently to a lower extent than in sub-Saharan Africa.
Corruption has a more deleterious impact on sub-Saharan African countries relative to the rest of the world. Turning next to corruption, the baseline model is estimated by replacing the governance indicator with two measures of corruption perceptions, CCI and CPI.17 The results for the core part of the model are roughly in line with those for the governance indicators (Table 2.2). The specifications also pass the tests for serial correlation and validity of instruments, as well as most robustness tests. Unlike the findings for governance, no supporting evidence was found for a statistically significant correlation between corruption perceptions and growth for the average country in the overall sample. However, both corruption perceptions indicators point to a stronger and negative correlation with economic performance in sub-Saharan Africa. As mentioned earlier, these correlations should not be interpreted literally as causation; however, a simulation shows that a 10-point improvement in corruption perceptions, or about one standard deviation in the sub-Saharan African sample—enough to move a sub-Saharan African country from the bottom quartile to the median of the sub-Saharan African distribution or bring the average sub-Saharan African country to the world average—would be associated with higher growth by 0.4–0.6 percentage points in the long run. These estimates are in the ballpark of other empirical investigations (for example, Gyimah-Brempong 2002; Gyimah-Brempong and de Camacho 2006; IMF 2018; Ugur and Dasgupta 2011).
A nonlinear correlation was found between governance/corruption and growth for both the overall and sub-Saharan African samples, with stronger correlations associated with weaker governance or higher corruption perceptions. The linear relationship between governance/corruption and growth has been disputed in the literature. For instance, the “grease in the wheels” view argues that some amount of corruption could increase bureaucratic efficiency and improve growth by mitigating red tape in developing countries. To test for a nonlinear correlation between governance/corruption perceptions and growth, a squared term was added for GOV in the vector of explanatory variables (see Hammadi and others 2019 for full estimation details). Next, the estimated coefficients were used to simulate the impact on growth from improving governance and reducing corruption. The finding, in this case, was that sub-Saharan African countries with governance (corruption perceptions) scores below (above) 60—about 75 percent of sub-Saharan Africa country-years in both samples—would benefit the most by addressing governance and corruption problems. As Figure 2.4 shows, the nonlinear models suggest that growth gains could be twice as large as the baseline for those sub-Saharan African countries with very low governance or high corruption scores (for example, at the 5th and 95th percentiles of the respective distributions).
Other dimensions of governance can also provide useful information—both about the specific transmission channels through which governance can aid growth and policy options to leverage these transmission channels. As part of this study, the governance indicators were examined—these individual subcomponents were interpreted as governance channels and were viewed as distinct but complementary transmission mechanisms of governance—and their correlation with growth was analyzed. Four key WGI components were considered: voice and accountability, government effectiveness, regulatory quality, and rule of law (see details in Hammadi and others 2019). Table 2.3 shows that the WGI’s four selected channels have a strong correlation with growth in sub-Saharan Africa. Using the estimated coefficients, the simulations used in this study indicate that the growth gains from a 10-point improvement in these specific components of governance range from ¾ of a percentage point for rule of law to about one percentage point for voice and accountability. These gains are nonadditive, but the overall impact could be higher due to potential complementarities across the different channels. Narrower variants of ICRG and WGI were also constructed by excluding corruption perceptions and a few other components to mitigate endogeneity and multicollinearity problems (for example, political stability, which to some extent is already captured by the measure of inflation used in this study). In a nutshell, these narrower components can be interpreted as proxies for the quality of political institutions and government regulations and policies. The empirical properties of the estimated models are somewhat similar to the baseline. Although the estimated correlations are less subject to concerns about endogeneity, their magnitudes are slightly smaller than in the baseline (see Hammadi and others 2019). In summary, unbundling governance channels may be important to better understand their empirical links with growth, and design tailored policy interventions, which can still yield significant growth dividends at the macroeconomic level.
Governance and Growth: Baseline Results1-2
The regressions are estimated using robust two-step system GMM estimator and include country and time fixed effects.
The instruments are the right-hand side controls (as shown above), lagged values of the dependent variable, and a dummy for resource intensity.
The serial correlation test is for second-order serial correlation and the Hansen test is for overidentifying restrictions.
The X2-test shows the test statistics and significance level from testing the null hypothesis that the coefficients on governance are jointly zero.
Governance and Growth: Baseline Results1-2
| Dependent variable: | (1) | (2) | (3) | (4) | (5) |
|---|---|---|---|---|---|
| Real GDP per capita growth | No governance | ICRG | WGI | ||
| No interaction | Interaction | No interaction | Interaction | ||
| Governance | 0.108*** | 0.107*** | 0.026 * | 0.036 ** | |
| Governance x SSA | 0.089 ** | 0.087 *** | |||
| Initial income per capita (log) | -2.031 *** | -2.435 *** | -2.765 *** | -1.929*** | -2.554 *** |
| Investment (percent of GDP) | 0.225 *** | 0.128 | 0.039 | 0.150*** | 0.134*** |
| Education (years) | 0.622 *** | 0.440 *** | 0.354 ** | 0.478 *** | 0.361 ** |
| High inflation dummy | -1.351” | -1.244** | -1.562*** | -1.901** | -1.760** |
| Change in terms of trade (percent) | 0.046 | 0.069 ** | 0.064 * | 0.077 ** | 0.080 ** |
| Constant | 13.49** | 13.27*** | 19.46*** | 13.27*** | 20.00 *** |
| Observations | 470 | 470 | 470 | 470 | 470 |
| Number of countries | 119 | 119 | 119 | 119 | 119 |
| Number of instruments3 | 14 | 15 | 17 | 15 | 17 |
| Serial correlation test (p-value)4 | 0.003 | 0.141 | 0.456 | 0.41 | 0.458 |
| Hansen test (p-value) | 0.031 | 0.327 | 0.283 | 0.22 | 0.255 |
| X2-test for HO: B, = 0 and B2 = 05 | 47.44 *** | 18.96*** | |||
The regressions are estimated using robust two-step system GMM estimator and include country and time fixed effects.
The instruments are the right-hand side controls (as shown above), lagged values of the dependent variable, and a dummy for resource intensity.
The serial correlation test is for second-order serial correlation and the Hansen test is for overidentifying restrictions.
The X2-test shows the test statistics and significance level from testing the null hypothesis that the coefficients on governance are jointly zero.
Governance and Growth: Baseline Results1-2
| Dependent variable: | (1) | (2) | (3) | (4) | (5) |
|---|---|---|---|---|---|
| Real GDP per capita growth | No governance | ICRG | WGI | ||
| No interaction | Interaction | No interaction | Interaction | ||
| Governance | 0.108*** | 0.107*** | 0.026 * | 0.036 ** | |
| Governance x SSA | 0.089 ** | 0.087 *** | |||
| Initial income per capita (log) | -2.031 *** | -2.435 *** | -2.765 *** | -1.929*** | -2.554 *** |
| Investment (percent of GDP) | 0.225 *** | 0.128 | 0.039 | 0.150*** | 0.134*** |
| Education (years) | 0.622 *** | 0.440 *** | 0.354 ** | 0.478 *** | 0.361 ** |
| High inflation dummy | -1.351” | -1.244** | -1.562*** | -1.901** | -1.760** |
| Change in terms of trade (percent) | 0.046 | 0.069 ** | 0.064 * | 0.077 ** | 0.080 ** |
| Constant | 13.49** | 13.27*** | 19.46*** | 13.27*** | 20.00 *** |
| Observations | 470 | 470 | 470 | 470 | 470 |
| Number of countries | 119 | 119 | 119 | 119 | 119 |
| Number of instruments3 | 14 | 15 | 17 | 15 | 17 |
| Serial correlation test (p-value)4 | 0.003 | 0.141 | 0.456 | 0.41 | 0.458 |
| Hansen test (p-value) | 0.031 | 0.327 | 0.283 | 0.22 | 0.255 |
| X2-test for HO: B, = 0 and B2 = 05 | 47.44 *** | 18.96*** | |||
The regressions are estimated using robust two-step system GMM estimator and include country and time fixed effects.
The instruments are the right-hand side controls (as shown above), lagged values of the dependent variable, and a dummy for resource intensity.
The serial correlation test is for second-order serial correlation and the Hansen test is for overidentifying restrictions.
The X2-test shows the test statistics and significance level from testing the null hypothesis that the coefficients on governance are jointly zero.
Corruption Perceptions and Growth: Baseline Results1,2
The regressions are estimated using robust two-step system GMM estimator and include country and time fixed effects.
The instruments are the right-hand side controls (as shown above), lagged values of the dependent variable, and a dummy for resource intensity.
The serial correlation test is for second-order serial correlation and the Hansen test is for overidentifying restrictions.
The X2-test shows the test statistics and significance level from testing the null hypothesis that the coefficients on governance are jointly zero.
Corruption Perceptions and Growth: Baseline Results1,2
| Dependent variable: | (1) | (2) | (3) | (4) | (5) | (6) | (7) |
|---|---|---|---|---|---|---|---|
| Real GDP per capita growth | WGI CCI | Transparency International CPI | |||||
| No corruption | No interaction | Interaction | Joint | No interaction | Interaction | Joint | |
| Corruption | -0.012 | -0.020 * | -0.002 | -0.002 | -0.012 | 0.0046 | |
| Corruption x SSA | -0.093 *** | -0.112*** | -0.090 *** | -0.106*** | |||
| Governance x LAC | -0.037 | -0.031 | |||||
| Governance x MENAAP | -0.073 ** | -0.057 ** | |||||
| Initial income per capita (log) | -1.654*** | -1.607 *** | -2.118*** | -1.942*** | -1.420*** | -1.863*** | -1.658 *** |
| Investment (percent of GDP) | 0.117 | 0.096 * | 0.096 * | 0.094 ** | 0.094 | 0.061 | 0.053 |
| Education (years) | 0.594 *** | 0.495 *** | 0.417*** | 0.270 * | 0.473 *** | 0.391 *** | 0.237* |
| High inflation dummy | -1.249* | -1.525* | -1.567** | -1.396* | -1.278* | -1.498 ** | -1.503** |
| Change in terms of trade (percent) | 0.013 | 0.020 | 0.020 | 0.022 | 0.013 | 0.019 | 0.021 |
| Constant | 12.43*** | 13.54*** | 19.41 *** | 17.97*** | 11.41*** | 17.60 *** | 16.04*** |
| Observations | 489 | 489 | 489 | 489 | 489 | 489 | 489 |
| Number of countries | 136 | 136 | 136 | 136 | 136 | 136 | 136 |
| Number of instruments3 | 14 | 15 | 17 | 21 | 15 | 17 | 21 |
| Serial correlation test (p-value)4 | 0.001 | 0.372 | 0.369 | 0.335 | 0.062 | 0.039 | 0.040 |
| Hansen test (p-value) | 0.053 | 0.279 | 0.250 | 0.203 | 0.149 | 0.235 | 0.228 |
| X2-test for HO: p, = 0 and p2 = 05 | 11.63 *** | 19.37*** | 8.71 ** | 16.5*** | |||
The regressions are estimated using robust two-step system GMM estimator and include country and time fixed effects.
The instruments are the right-hand side controls (as shown above), lagged values of the dependent variable, and a dummy for resource intensity.
The serial correlation test is for second-order serial correlation and the Hansen test is for overidentifying restrictions.
The X2-test shows the test statistics and significance level from testing the null hypothesis that the coefficients on governance are jointly zero.
Corruption Perceptions and Growth: Baseline Results1,2
| Dependent variable: | (1) | (2) | (3) | (4) | (5) | (6) | (7) |
|---|---|---|---|---|---|---|---|
| Real GDP per capita growth | WGI CCI | Transparency International CPI | |||||
| No corruption | No interaction | Interaction | Joint | No interaction | Interaction | Joint | |
| Corruption | -0.012 | -0.020 * | -0.002 | -0.002 | -0.012 | 0.0046 | |
| Corruption x SSA | -0.093 *** | -0.112*** | -0.090 *** | -0.106*** | |||
| Governance x LAC | -0.037 | -0.031 | |||||
| Governance x MENAAP | -0.073 ** | -0.057 ** | |||||
| Initial income per capita (log) | -1.654*** | -1.607 *** | -2.118*** | -1.942*** | -1.420*** | -1.863*** | -1.658 *** |
| Investment (percent of GDP) | 0.117 | 0.096 * | 0.096 * | 0.094 ** | 0.094 | 0.061 | 0.053 |
| Education (years) | 0.594 *** | 0.495 *** | 0.417*** | 0.270 * | 0.473 *** | 0.391 *** | 0.237* |
| High inflation dummy | -1.249* | -1.525* | -1.567** | -1.396* | -1.278* | -1.498 ** | -1.503** |
| Change in terms of trade (percent) | 0.013 | 0.020 | 0.020 | 0.022 | 0.013 | 0.019 | 0.021 |
| Constant | 12.43*** | 13.54*** | 19.41 *** | 17.97*** | 11.41*** | 17.60 *** | 16.04*** |
| Observations | 489 | 489 | 489 | 489 | 489 | 489 | 489 |
| Number of countries | 136 | 136 | 136 | 136 | 136 | 136 | 136 |
| Number of instruments3 | 14 | 15 | 17 | 21 | 15 | 17 | 21 |
| Serial correlation test (p-value)4 | 0.001 | 0.372 | 0.369 | 0.335 | 0.062 | 0.039 | 0.040 |
| Hansen test (p-value) | 0.053 | 0.279 | 0.250 | 0.203 | 0.149 | 0.235 | 0.228 |
| X2-test for HO: p, = 0 and p2 = 05 | 11.63 *** | 19.37*** | 8.71 ** | 16.5*** | |||
The regressions are estimated using robust two-step system GMM estimator and include country and time fixed effects.
The instruments are the right-hand side controls (as shown above), lagged values of the dependent variable, and a dummy for resource intensity.
The serial correlation test is for second-order serial correlation and the Hansen test is for overidentifying restrictions.
The X2-test shows the test statistics and significance level from testing the null hypothesis that the coefficients on governance are jointly zero.
Overall Effect of Governance and Corruption Perceptions on Growth
Source: Authors’ estimates.Note: This figure shows the overall marginal effect (in percentage points) of corruption (right-hand side, using CCI) and governance (left-hand side, using ICRG) on GDP per capita growth. In each chart, the horizontal axis measures the normalized governance or corruption indicator, whereas the vertical axis measures the impact on growth. The negatively sloped solid lines are the average estimates for the marginal effect for the nonlinear model, whereas the horizontal dotted lines are the point estimates for the baseline linear model. The shaded areas are +/– two standard deviations around the average estimates.Overall Effect of Governance and Corruption Perceptions on Growth
Source: Authors’ estimates.Note: This figure shows the overall marginal effect (in percentage points) of corruption (right-hand side, using CCI) and governance (left-hand side, using ICRG) on GDP per capita growth. In each chart, the horizontal axis measures the normalized governance or corruption indicator, whereas the vertical axis measures the impact on growth. The negatively sloped solid lines are the average estimates for the marginal effect for the nonlinear model, whereas the horizontal dotted lines are the point estimates for the baseline linear model. The shaded areas are +/– two standard deviations around the average estimates.Overall Effect of Governance and Corruption Perceptions on Growth
Source: Authors’ estimates.Note: This figure shows the overall marginal effect (in percentage points) of corruption (right-hand side, using CCI) and governance (left-hand side, using ICRG) on GDP per capita growth. In each chart, the horizontal axis measures the normalized governance or corruption indicator, whereas the vertical axis measures the impact on growth. The negatively sloped solid lines are the average estimates for the marginal effect for the nonlinear model, whereas the horizontal dotted lines are the point estimates for the baseline linear model. The shaded areas are +/– two standard deviations around the average estimates.Taken together, the evidence gathered in this chapter suggests that the correlation between governance/corruption and growth in sub-Saharan Africa could be stronger than in other regions. While not entirely conclusive, the nonlinear models suggest that very low levels of governance or high corruption perceptions are more detrimental to growth than modest levels. In addition, there could be region-specific factors—for example, very weak or ineffective institutions and capacity, latent rent-seeking in resource-rich sectors, or perhaps the fact that corruption has become entrenched in everyday expecta-tions—that may partly explain the amplified impact on growth in sub-Saharan Africa.
Overall, this evidence for sub-Saharan Africa suggests that poor governance and corruption are both acute and endemic problems in the region. Weak institutions and resource intensity often coexist with high incidence of corruption in sub-Saharan Africa. The large economic rents generated by resource-rich sectors such as oil—often controlled by state-owned companies subject to political interference—can expose resource-rich countries to higher levels of corruption, especially when institutions are weak (OECD 2014; 2016).
Governance and Growth: Channels1,2
The regressions are estimated using robust two-step system GMM estimator and include country and time fixed effects.
The instruments are the right-hand side controls (as shown above), lagged values of the dependent variable, and a dummy for resource intensity.
The serial correlation test is for second-order serial correlation and the Hansen test is for overidentifying restrictions.
The X2-test shows the test statistics and significance level from testing the null hypothesis that the coefficients on governance are jointly zero.
Governance and Growth: Channels1,2
| Dependent variable: | (1) | (2) | (3) | (4) | (5) | (6) | (7) | (8) |
|---|---|---|---|---|---|---|---|---|
| Real GDP per capita growth | Voice and Accountability | Government Effectiveness | Regulatory Quality | Rule of Law | ||||
| Governance | 0.012 | 0.004 | 0.036 | 0.046 *** | 0.034 | 0.036 * | 0.006 | 0.017 |
| Governance x SSA | 0.099 *** | 0.074 *** | 0.085 *** | 0.072 *** | ||||
| Initial income per capita (log) | -1.713*** | -2.095 *** | -2.079 *** | -2.707 *** | -1.973*** | -2.489 *** | -1.705*** | -2.296 *** |
| Investment (percent of GDP) | 0.147** | 0.119** | 0.155*** | 0.148*** | 0.159** | 0.145 *** | 0.145 ** | 0.129** |
| Education (years) | 0.504 *** | 0.495 *** | 0.464 *** | 0.338 * | 0.464 *** | 0.380 ** | 0.534 *** | 0.434 ** |
| High inflation dummy | -2.062 ** | -2.130** | -1.914** | -1.835** | -1.854* | -1.796** | -2.062 ** | -2.014** |
| Change in terms of trade (percent) | 0.073 ** | 0.070 ** | 0.081 ** | 0.086 ** | 0.079 ** | 0.082 ** | 0.074 ** | 0.080 ** |
| Constant | 11.98 *** | 17.15*** | 14.16*** | 20.74 *** | 12.99*** | 18.84*** | 12.19*** | 18.44*** |
| Observations | 470 | 470 | 470 | 470 | 470 | 470 | 470 | 470 |
| Number of countries | 119 | 119 | 119 | 119 | 119 | 119 | 119 | 119 |
| Number of instruments3 | 15 | 17 | 15 | 17 | 15 | 17 | 15 | 17 |
| Serial correlation test (p-value)4 | 0.430 | 0.570 | 0.420 | 0.510 | 0.370 | 0.370 | 0.430 | 0.470 |
| Hansen test (p-value) | 0.170 | 0.200 | 0.260 | 0.280 | 0.210 | 0.210 | 0.160 | 0.200 |
| X2-test for H0: p, = 0 and p2 = 05 | 15.75*** | 23.91 *** | 17.41 *** | 15.15*** | ||||
The regressions are estimated using robust two-step system GMM estimator and include country and time fixed effects.
The instruments are the right-hand side controls (as shown above), lagged values of the dependent variable, and a dummy for resource intensity.
The serial correlation test is for second-order serial correlation and the Hansen test is for overidentifying restrictions.
The X2-test shows the test statistics and significance level from testing the null hypothesis that the coefficients on governance are jointly zero.
Governance and Growth: Channels1,2
| Dependent variable: | (1) | (2) | (3) | (4) | (5) | (6) | (7) | (8) |
|---|---|---|---|---|---|---|---|---|
| Real GDP per capita growth | Voice and Accountability | Government Effectiveness | Regulatory Quality | Rule of Law | ||||
| Governance | 0.012 | 0.004 | 0.036 | 0.046 *** | 0.034 | 0.036 * | 0.006 | 0.017 |
| Governance x SSA | 0.099 *** | 0.074 *** | 0.085 *** | 0.072 *** | ||||
| Initial income per capita (log) | -1.713*** | -2.095 *** | -2.079 *** | -2.707 *** | -1.973*** | -2.489 *** | -1.705*** | -2.296 *** |
| Investment (percent of GDP) | 0.147** | 0.119** | 0.155*** | 0.148*** | 0.159** | 0.145 *** | 0.145 ** | 0.129** |
| Education (years) | 0.504 *** | 0.495 *** | 0.464 *** | 0.338 * | 0.464 *** | 0.380 ** | 0.534 *** | 0.434 ** |
| High inflation dummy | -2.062 ** | -2.130** | -1.914** | -1.835** | -1.854* | -1.796** | -2.062 ** | -2.014** |
| Change in terms of trade (percent) | 0.073 ** | 0.070 ** | 0.081 ** | 0.086 ** | 0.079 ** | 0.082 ** | 0.074 ** | 0.080 ** |
| Constant | 11.98 *** | 17.15*** | 14.16*** | 20.74 *** | 12.99*** | 18.84*** | 12.19*** | 18.44*** |
| Observations | 470 | 470 | 470 | 470 | 470 | 470 | 470 | 470 |
| Number of countries | 119 | 119 | 119 | 119 | 119 | 119 | 119 | 119 |
| Number of instruments3 | 15 | 17 | 15 | 17 | 15 | 17 | 15 | 17 |
| Serial correlation test (p-value)4 | 0.430 | 0.570 | 0.420 | 0.510 | 0.370 | 0.370 | 0.430 | 0.470 |
| Hansen test (p-value) | 0.170 | 0.200 | 0.260 | 0.280 | 0.210 | 0.210 | 0.160 | 0.200 |
| X2-test for H0: p, = 0 and p2 = 05 | 15.75*** | 23.91 *** | 17.41 *** | 15.15*** | ||||
The regressions are estimated using robust two-step system GMM estimator and include country and time fixed effects.
The instruments are the right-hand side controls (as shown above), lagged values of the dependent variable, and a dummy for resource intensity.
The serial correlation test is for second-order serial correlation and the Hansen test is for overidentifying restrictions.
The X2-test shows the test statistics and significance level from testing the null hypothesis that the coefficients on governance are jointly zero.
Policy Implications
The findings outlined in this chapter emphasize the importance of buttressing the fight against corruption and striving to improve governance. The nonlinear-ities in the relationship between corruption and growth suggest there is no one-size-fits-all approach (for example, Klitgaard 1988; Mungiu-Pippidi and Johnston 2017). Improving governance is not without its challenges, as the beneficiaries of the proceeds of corruption will often fight back. It is thus a complex, and likely drawn-out, battle reflecting the interaction between the various players—government, institutions, civil society, media, and the private sector. Development partners and multinationals also have an important role on the supply side. Strong political commitment is thus a sine qua non for success.
From an economic perspective, there are some basic principles that apply across countries and can strengthen governance. These include improving regulatory quality and government effectiveness and strengthening fiscal institutions. Indeed, the successful experiences of countries like Botswana, Chile, Estonia, and Georgia suggest that multiple factors may have contributed to their success, including: political will; measures to reduce corruption opportunities (for example, lowering red tape and trade barriers); measures to increase constraints on corrupt behavior (for example, independent judicial system and pressure from civil society); and improved fiscal institutions (greater fiscal transparency and controls).
While institutional reforms take time, in most cases enforcing compliance with existing laws and regulations can be the first step in the right direction. Empowering and capacitating anticorruption institutions will improve their prosecution capability and bridge the gap between public opinion and the courts of law. Corruption prosecution cases often fail, as governments may lack the financial resources or the legal capacity to successfully investigate and present convincing evidence of wrongdoing to the courts. Further improving the rule of law and building checks and balances, particularly by improving the governance structure of state-owned enterprises, will also help. This includes implementing measures related to customer due diligence, beneficial ownership, asset declarations, and politically exposed persons, measures that can support an effective framework for AML/CFT. As shown in Chapter 15, digitalization is also opening new possibilities by facilitating taxation, improving spending efficiency, and making procurement more transparent, inclusive, and efficient.
Conclusion
In this chapter, a strong correlation was found between governance/corruption perceptions and growth in sub-Saharan Africa, with similar results for other regions (for example, LAC, MENAAP) that are also perceived to experience more acute governance and corruption problems. The results show that the governance dividend for the average sub-Saharan Africa country would be two to three times larger than the rest of the world. For instance, bringing sub-Saharan Africa’s governance to the world average could increase GDP per capita by an estimated 1 to 2 percentage points per year. This is the case because growth dividends may be larger for countries that are perceived to have severe governance/corruption problems, but less so if such problems are perceived as moderate or small. In other words, the payoffs associated with reforms to improve governance are likely to be nonlinear, with a bigger impact when there are major weaknesses and less substantial payoffs after countries have implemented substantial reforms. The baseline results are robust to several tests that attempt to mitigate potential endogeneity problems in the relationship between governance/corruption and growth. The findings remain largely robust to various specifications and sample choices.
As sub-Saharan African countries consider their policy mix to reignite post-pandemic recovery and growth, potentially significant economic gains can be achieved by improving governance and reducing corruption. Strengthening governance and fighting corruption are not easy tasks—and the process is often time-consuming and requires considerable political effort—but they are worth pursuing given the potentially large payoffs.
Annex 2.1. Data and Baseline Results
Summary of Data and Data Sources
The controls in vector X in Box 2.1, equation (1), are measured as follows: (Annex Table 2.1.1):
Initial GDP per capita: Real GDP per capita Purchasing Power Parity (in log) in the year immediately before each five-year period, to control for income convergence.
Gross capital formation: Total investment, public and private (percent of GDP), to capture the contribution of capital accumulation to growth. To mitigate endogeneity problems, investment is measured in the first year of each five-year period.
Level of education: Per capita years of secondary and tertiary schooling for the population aged 15 and above, to capture the contribution of human capital.
Dummy variable for high inflation: This takes value 1 if inflation is larger than 15 percent—at the top quartile of the sample distribution—and zero otherwise. It proxies for the quality of macroeconomic policies and macro-economic volatility.
Terms of trade: Percentage change in terms of trade, to reflect external shocks.
In the baseline regressions, two aggregate indicators of governance and two indicators of corruption perceptions are used (Annex Table 2.1.1):
ICRG: This indicator covers several aspects of governance, including government stability, internal and external conflicts, corruption, law and order, ethnic tensions, democratic accountability, and bureaucracy quality. It covers a large time span (data are available since 1984) but only covers two-thirds of sub-Saharan Africa countries.
WGI: For the purposes of the regressions in this chapter, this is constructed as the simple sum of the six WGI from Kaufmann and Kraay (Kaufmann, Kraay, and Mastruzzi 2010): voice and accountability, political stability and absence of violence and terrorism, government effectiveness, regulatory quality, rule of law, and control of corruption. It covers all sub-Saharan Africa countries and data are available since 1996.
CCI: This indicator of corruption perceptions is one component of WGI and data are also available since 1996.
CPI: This indicator covers data available since 1995, starting with a relatively small country coverage that was gradually expanded to cover all sub-Saharan Africa countries.
Despite differences in methodology and sample, these indicators are highly correlated. While subjective and subject to various caveats (see “Measurement, Stylized Facts, and Channels for a Governance Dividend”), Hamilton and Hammer (2018) argue that the corruption perceptions measured by CCI and CPI are sufficiently comprehensive to capture all elements of corruption and hence are a good starting point of empirical analyses.
Baseline Results for Sub-Saharan Africa
All coefficients on the core controls have the expected sign and almost all are statistically significant (see Table 2.1). The coefficient on the interaction term for sub-Saharan Africa (β2) has the positive expected sign and is also statistically significant. The specifications pass the tests for absence of serial correlation and validity of instruments. The null hypothesis that the coefficient β1 and β2 are jointly zero is also rejected.
These correlations should not be interpreted as causal effects. But mechanically, a 10-point improvement in governance—equivalent to one standard deviation in the sub-Saharan Africa sample and enough to move a sub-Saharan Africa country from the bottom quartile to the median of the sub-Saharan Africa distribution or bring the average sub-Saharan Africa country to the world average— would be associated with higher growth by between 1¼ percentage point (WGI) and 2 percentage points (ICRG). This impact is about thrice and twice larger than that for the overall sample, respectively. As mentioned earlier, some of the covariates like education and inflation also capture some of the indirect effects of governance on growth. Therefore, the total impact of governance on growth— including the indirect effects through these variables—is likely larger than β1+ β2.
Controlling for endogeneity problems, a strong correlation is still found between governance and growth in sub-Saharan Africa. The model also survives several robustness tests, including the use of alternative estimation methods (for example, difference GMM, OLS), inclusion of additional controls, and choice of the averaging window (see Hammadi and others 2019).
Summary of Variables and Data Sources
Observations with average annual changes of ±20 percent or more over any five-year period are excluded.
Observations for the five-year period 2011–15 were obtained by extrapolating Barro-Lee data.
The dummy variable takes value 1 if average annual inflation exceeds 15 percent over a given five-year period, and zero otherwise.
Summary of Variables and Data Sources
| Variable | Data Description | Data Sources | Sample Period | Expected Sign |
|---|---|---|---|---|
| Dependent Variable | ||||
| Growth1 | GDP per Capita Growth Rate (Percent) | Penn World Tables 9.0, WEO | 1980–15 | |
| Right-Hand-Side Variables | ||||
| GDP per Capita | Measured at Purchasing Power Parity (PPP), log | Penn World Tables 9.0, WEO | 1980–14 | (–) |
| Investment | Gross capital formation, percent of GDP | Penn World Tables 9.0 | 1980–14 | (+) |
| Education2 | Secondary and tertiary education, years | Barro-Lee database | 1980–15 | (+) |
| Inflation3 | 1 if inflation > 15 percent, 0 otherwise | WEO | 1980–15 | (–) |
| Terms of Trade | Annual percent change | WEO, WDI | 1980–15 | (+) |
| ICRG | Governance indicator, 0–100 | ICRG | 1984–15 | (+) |
| WGI | Governance indicator, 0–100 | Kaufmann and Kraay | 1996–15 | (+) |
| CCI | Corruption perceptions, 0–100 | Kaufmann and Kraay | 1996–15 | (–) |
| CPI | Corruption perceptions, 0–100 | Transparency International | 1995–15 | (–) |
Observations with average annual changes of ±20 percent or more over any five-year period are excluded.
Observations for the five-year period 2011–15 were obtained by extrapolating Barro-Lee data.
The dummy variable takes value 1 if average annual inflation exceeds 15 percent over a given five-year period, and zero otherwise.
Summary of Variables and Data Sources
| Variable | Data Description | Data Sources | Sample Period | Expected Sign |
|---|---|---|---|---|
| Dependent Variable | ||||
| Growth1 | GDP per Capita Growth Rate (Percent) | Penn World Tables 9.0, WEO | 1980–15 | |
| Right-Hand-Side Variables | ||||
| GDP per Capita | Measured at Purchasing Power Parity (PPP), log | Penn World Tables 9.0, WEO | 1980–14 | (–) |
| Investment | Gross capital formation, percent of GDP | Penn World Tables 9.0 | 1980–14 | (+) |
| Education2 | Secondary and tertiary education, years | Barro-Lee database | 1980–15 | (+) |
| Inflation3 | 1 if inflation > 15 percent, 0 otherwise | WEO | 1980–15 | (–) |
| Terms of Trade | Annual percent change | WEO, WDI | 1980–15 | (+) |
| ICRG | Governance indicator, 0–100 | ICRG | 1984–15 | (+) |
| WGI | Governance indicator, 0–100 | Kaufmann and Kraay | 1996–15 | (+) |
| CCI | Corruption perceptions, 0–100 | Kaufmann and Kraay | 1996–15 | (–) |
| CPI | Corruption perceptions, 0–100 | Transparency International | 1995–15 | (–) |
Observations with average annual changes of ±20 percent or more over any five-year period are excluded.
Observations for the five-year period 2011–15 were obtained by extrapolating Barro-Lee data.
The dummy variable takes value 1 if average annual inflation exceeds 15 percent over a given five-year period, and zero otherwise.
References
Baum, Anja, S. Gupta, E. Kimani, and S. Tapsoba. 2017. “Corruption, Taxes and Compliance.” IMF Working Paper 17/255, International Monetary Fund, Washington, DC.
Blundell, Richard, and S. Bond. 1998. “Initial Conditions and Moment Restrictions in Dynamic Panel Data Models.” Journal of Econometrics 87 (1): 115–143.
Cerqueti, Roy, R. Coppier, and G. Piga. 2012. “Corruption, Growth and Ethnic Fractionalization: A Theoretical Model.” Journal of Economics 106 (2): 153–181.
Collier, Paul, and J. Gunning. 1999. “Explaining African Economic Performance.” Journal of Economic Literature 37 (1): 64–111.
Giavazzi, Francesco, and G. Tabellini. 2005. “Economic and Political Liberalizations.” Journal of Monetary Economics 52 (7): 1297–330.
Gupta, Sanjeev, H. Davoodi, and R. Alonso-Terme. 2002. “Does Corruption Affect Income Inequality and Poverty?” Economics of Governance 3 (1): 23–45.
Gyimah-Brempong, K. 2002. “Corruption, Economic Growth, and Income Inequality in Africa.” Economics of Governance 3 (3): 183–209.
Gyimah-Brempong, Kwabena, and S. M. de Camacho. 2006. “Corruption, Growth, and Income Distribution: Are There Regional Differences?” Economics of Governance 7 (3): 245–269.
Hamilton, Alexander, and C. Hammer. 2018. “Can We Measure the Power of the Grabbing Hand? A Comparative Analysis of Different Indicators of Corruption.” Policy Research Working Paper 8299, World Bank, Washington, DC.
Hammadi, Amine, M. Mills, N. Sobrinho, V. Thakoor, and R. Velloso. 2019. “A Governance Dividend for Sub-Saharan Africa?” IMF Working Paper 19/1, International Monetary Fund, Washington, DC.
International Monetary Fund (IMF). 2016. “Corruption: Costs and Mitigating Strategies.” IMF Staff Discussion Note 16/05, Washington, DC.
International Monetary Fund (IMF). 2017. “The Role of the Fund in Governance Issues— Review of the Guidance Note—Preliminary Considerations.” IMF Policy Paper, Washington, DC.
International Monetary Fund (IMF). 2018. “Review of the 1997 Guidance Note on Governance—a Proposed Framework for Enhanced Fund Engagement.” IMF Policy Paper, Washington, DC.
Kaufmann, Daniel, A. Kraay, and M. Mastruzzi. 2010. “The Worldwide Governance Indicators: Methodology and Analytical Issues.” Policy Research Working Paper 5430, World Bank, Washington, DC.
Klitgaard, Robert. 1988. Controlling Corruption. Berkeley: University of California Press.
Mauro, Paolo. 1995. “Corruption and Growth.” Quarterly Journal of Economics 110(3): 681–712.
Mungiu-Pippidi, Alina, and M. Johnston. 2017. Transitions to Good Governance. Cheltenham, UK: Edward Elgar.
Organisation for Economic Co-operation and Development (OECD). 2014. OECD Foreign Bribery Report: An Analysis of the Crime of Bribery of Foreign Public Officials. Paris: OECD Publishing.
Organisation for Economic Co-operation and Development (OECD). 2016. Corruption in the Extractive Value Chain: Typology of Risks, Mitigation Measures, and Incentives. Paris: OECD Publishing.
Roodman, David. 2009. “How to Do Xtabond2: An Introduction to Difference and System GMM in Stata.” Stata Journal 9 (1): 86–136.
Saha, Shrabani, and R. Gounder. 2013. “Corruption and Economic Development Nexus: Variations across Income Levels in an Income-Linear Framework.” Economic Modelling 31: 70–79.
Saha, Shrabani, G. Mallik, and D. Vortelinos. 2017. “Does Corruption Facilitate Growth? A Cross-National Study in a Non-linear Framework.” South Asia Journal of Macroeconomics and Public Finance 6 (2): 1–16.
Shleifer, Andrei, and R. Vishny. 1993. “Corruption.” Quarterly Journal of Economics 108 (3): 599–617.
Tanzi, Vito, and H. Davoodi. 1997. “Corruption, Public Investment, and Growth.” IMF Working Paper 97/139, International Monetary Fund, Washington, DC.
Ugur, Mehmet, and N. Dasgupta. 2011. Evidence on the Economic Growth Impacts of Corruption in Low-Income Countries and Beyond: A Systematic Review. London: EPPI-Centre, Social Science Research Unit, Institute of Education, University of London.
This chapter draws heavily on joint work done with Amine Hammadi and Ricardo Velloso as presented in Hammadi and others (2019), which was part of the IMF’s broader effort to study the economic and policy implications of the quality of governance worldwide, especially in developing countries.
We view corruption as one important manifestation of governance weakness, hence the significant emphasis on corruption in this chapter.
Similar results hold across a range of corruption and governance indicators. Corruption is defined as “the abuse of public office for private gain.” Governance is defined as follows: “Institutions, mechanisms, and practices through which government power is exercised in a country, including for the management of public resources and regulation of the economy. This includes processes at the country level, including institutions-level structural arrangements” (IMF 2017). Governance indicators are normalized to range from 0 (worst) to 100 (best). Corruption perceptions indicators are also normalized to range from 0 (best) to 100 (worst). Given the degree of uncertainty around point estimates of governance and corruption indicators, we report confidence intervals and standard errors instead. We also note that these indicators, especially of corruption, are based on perceptions and may not be readily comparable across countries, regions, or time.
See, for example, Mauro (1995); Shleifer and Vishny (1993); and Ugur and Dasgupta (2011).
See, for example, Cerqueti, Coppier, and Piga (2012); Saha and Gounder (2013); and Saha, Mallik, and Vortelinos (2017).
See, for instance, Kaufmann, Kraay, and Mastruzzi (2010).
See Collier and Gunning (1999); Gyimah-Brempong (2002); and Gyimah-Brempong and de Cama-cho (2006).
See, for example, Baum and others (2017) and Tanzi and Davoodi (1997).
The strong correlation between corruption perceptions and overall measures of governance also arises by construction as corruption perceptions are included in most aggregate governance indices.
See, for example, Baum and others (2017).
See, for example, Tanzi and Davoodi (1997).
See, for example, Mauro (1995) and (Tanzi and Davoodi 1997).
See, for example, Gupta, Davoodi, and Alonso-Terme (2002).
One standard deviation in the sub-Saharan African sample for both governance and corruption perceptions is equivalent to 10 percentage points if all scores are normalized to 0–100.
The corruption index from ICRG was also tested, but the correlation was estimated less reliably, reflecting data availability (for example, narrower country coverage than CCI and CPI in the first half of the sample).
