# The Impact of Corruption and Shadow Economy on the Economic and Sustainable Development. Do They “Sand the Wheels” or “Grease the Wheels”?

^{1}

^{2}

^{3}

^{4}

^{5}

^{*}

## Abstract

**:**

## 1. Introduction

## 2. Literature Review

## 3. Methodology and Data

#### 3.1. Working Hypotheses and Research Questions

**Hypothesis**

**1.**

**Hypothesis**

**2.**

**Hypothesis**

**3.**

**Hypothesis**

**4.**

**Research question****1.**- How does the impact of corruption upon economic development differ among high-income and low-income countries?
**Research question****2.**- How does the impact of corruption upon sustainable development differ among high-income and low-income countries?
**Research question****3.**- How does the impact of the shadow economy upon economic development differ among high-income and low-income countries?
**Research question****4.**- How does the impact of the shadow economy upon sustainable development differ among high-income and low-income countries?

#### 3.2. Defining and Measuring the Variables

#### 3.2.1. Defining and Measuring the Dependent Variables: Corruption and Shadow Economy

**a**. to avoid paying income, value-added, or other taxes,

**b**. to avoid paying social security contributions,

**c**. to avoid having to meet certain legal labour market standards, such as minimum wages, maximum working hours, and safety standards, and

**d**. to avoid complying with certain administrative procedures, such as completing statistical questionnaires or other administrative forms”. We rely on this narrow definition of the shadow economy that refers only to activities that are concealed from the government, rather than to those that are illegal (activities that produce branded goods illegally, drug trafficking, prostitution, loan sharking, illegal gambling, hiring of illegal immigrants, hidden income, and tax fraud). We measure the shadow economy using data provided by Medina and Schneider [35] for the 2005–2015 time period, throughout which it is calculated as a percentage of the official GDP.

#### 3.2.2. Defining Independent Variables: Economic and Sustainable Development

- (
**a**) - The human development index (HDI) reflects a final criterion for evaluating the development of a country, not just for economic growth. It is a summary measure of the average achievement in three key dimensions of human development: a long and healthy life, being aware of the cause and having a decent standard of living, according to UNDP Human Development Reports [45]. The HDI indicator is used in many research papers as a proxy for sustainable development [22,46].
- (
**b**) - Environmental performance index (EPI) which provides a national scale of how close countries are to meeting the established environmental policy goals. The EPI thus offers a scorecard that highlights leaders and laggards in environmental performance, gives insight on best practices, and provides guidance for countries that aspire to be leaders in sustainability [47]. Environmental Performance Index (EPI) ranks 180 countries on 24 performance indicators across ten issue categories covering environmental health and ecosystem vitality. It is used in many studies as a proxy for environmental performances [48,49,50]. The environmental indicators comprised by EPI are focused on two objectives: on the one hand, a reduction in environmental stresses to human health and, on the other hand, the protection of ecosystems and natural resources [48,49].

#### 3.3. Sample and Data

#### 3.4. Descriptive Analysis

## 4. Results and Discussions

^{−0.0232}= 0.97706, or a 2.2932% decrease. So, the lower the corruption ranking (increase in the rank position), the lower the per capita GDP of countries, with 2.2932% lower for each one rank increase. The explicative power of this first Pooled OLS model is of 57.04%, rather powerful. However, using the FEM method we have a positive coefficient of COR of 0.0023, significant at an only 10% threshold. This means that each one-unit increase in COR multiplies the expected value of GDP by e

^{0.0023}=1.002302 or a 0.23% increase. When we use REM method, the coefficient of corruption is also negative (like in the case of Pooled OLS) but is not statistically significant.

^{−0.0111}= 0.98896 or a 1.1038% decrease. The Adj R

^{2}is of 0.3107 on the Pooled OLS estimation technique.

^{−0.0126}=0.98747 or a 1.26% decrease. The Adj R

^{2}for this Pooled OLS model is of 21.11%.

^{0.0024}=1.002402 or a 0.2402% increase. Thus, we get supplementary evidence for a positive impact of corruption upon the level of development faced by low-income countries.

^{2}for the Pooled OLS model is of 0.3107) compared to low-income countries (the Adj R

^{2}for the Pooled OLS model is of only 0.2111). We see that in low-income countries, the negative impact of corruption is diminished compared to high-income countries, and we even find positive effects of corruption upon the level of state development (through the use of FEM). Thus, we may sum up on a rather mixed role of the size of corruption on the economic development of low-income countries while for the high-income countries’ subsample we have clear evidence of a negative relationship. Thus, for low-income countries with weaker governance than high-income countries, corruption may help firms avoid government regulations and therefore their profitability enhances. These positive influences are also validated by the literature research among the supporters of the “grease the wheels” idea [4,6,8,26].

^{−0.0887}=0.91512, or an 8.4879% decrease. So, the increase in the shadow economy determines the decrease in economic development. The explicative power of this Pooled OLS model is strong, having an R

^{2}of 47.82%. Moreover, the panel diagnosis tests point towards FEM as the optimal estimation technique. For FEM, the coefficient of SE is of −0.0547. Basically, each one-unit increase in SE multiplies the expected value of GDP by e

^{−0.0547}=0.9467 or a 5.32% decrease. Thus, all the run tests applied for the entire sample document a negative and significant coefficient of SE in relation to LogGDP. Therefore, our finding supports the idea that the higher the shadow economy, the lower the economic development and thus our hypothesis (Hypothesis 3. The higher the level of shadow economy the lower the level of economic development) is accepted. Our findings are in line with the “sand the wheels” literature strand regarding the negative influence of the shadow economy on economic development [24,25]. Underground activities come along with tax payments avoidance further decreasing the income of the state. The state would need these incomes in order to support its public investments, to cover the expenses of public institutions, to support the development of the national economy and to cover healthcare, education and citizens’ protection expenditures. All these may finally hamper productivity and growth [25]. A decrease in the state’s income will lead to a reduction of its financial power, which is necessary in order to ensure the normal functioning of state institutions and authorities. The effects of fiscal evasion will be felt both by the honest taxpayers and by the ones that avoid these compulsory payments.

^{−0.0444}=0.95657, or a 4.3428% decrease. The Adj R

^{2}is of 0.3251 on the Pooled OLS estimation technique, so SE explicates 32.51% of economic development for the high-income countries’ subsample. Furthermore, the Hausman test points towards REM as the optimal estimation technique in the regression of LogGDP against SE for high-income countries.

^{−0.0375}=0.96319, or a 3.68% decrease. So, the increase in the shadow economy actually determines the decrease in economic development. The Adj R

^{2}for this Pooled OLS model is of 12.88%. Moreover, the panel diagnosis tests point towards REM as the optimal estimation technique. For REM, the coefficient of SE is of −0.0551, significant at a 5% threshold. Basically, each one-unit increase in SE multiplies the expected value of GDP by e

^{−0.0551}=0.94639 or a 5.36% decrease. The higher the weight of the shadow economy, the lower the economic development.

^{2}for the Pooled OLS model is of 0.3251) compared to low-income countries (Adj R

^{2}for the Pooled OLS model is of 0.1288). In other words, for high-income countries, the negative effects of shadow activities upon economic development are higher than in the case of low-income countries. Our results are somehow in line with the findings of Williams [52], Zaman and Goschin [11], and Ruzek [27] who document that informal activities represent an important buffer for solving many economic problems especially for the less developed countries. Thus, if in developed countries the informal economy can be a choice, in less developed countries it is out of economic necessity [27,52]. Thus, for low-income countries, the negative effects of the shadow economy are significantly diminished, which is in line with our findings.

^{2}is of 0.5826 on the Pooled OLS estimation technique. Although the Hausman test points towards FEM as the optimal estimation technique, the positive coefficient of COR is not significant on FEM. The right part of Table A3b deals with the subsample of low-income countries. In terms of effects of COR on HDI, we obtain a 0.0015 decrease in HDI for a one-unit increase in COR, significant at 1%. The Adj R

^{2}for this Pooled OLS model is of 19.54%. Moreover, the panel diagnosis tests point towards FEM as the optimal estimation technique for COR, but the positive coefficient of COR is not significant.

^{2}for the Pooled OLS model is of 0.5826) compared to low-income countries (the Adj R

^{2}for the Pooled OLS model is of 0.1954). Accordingly, we may see that in low-income countries, the negative impact of corruption is highly diminished compared to high-income countries.

^{2}for the Pooled OLS model is of 0.4046) compared to low-income countries (the Adj R

^{2}for the Pooled OLS model is of 0.106). We may see that in low-income countries the negative impact of shadow economy upon sustainable development (expressed by HDI) is highly diminished. Our results are in line with the findings of Williams [52], Zaman and Goschin [11] and Ruzek [27] who document the positive role held by shadow activities for solving many economic problems especially in developing countries, and therefore for finding the proper channels to increase the level of sustainable development.

^{2}is of 0.2178 on the Pooled OLS estimation technique. Although the Hausman test points towards FEM as the optimal estimation technique, the positive coefficient of COR is not significant on FEM. For REM, the coefficient of COR is still negative and significant. In Models (2), for high-income countries, we obtain a −0.7544 coefficient of SE, significant at a 1% threshold, so each one-unit increase of SE decreases EPI by 0.75 units. The Adj R

^{2}is of 0.2257 on the Pooled OLS estimation technique, so we obtain that 22.57% of the amount of variance in EPI is explained by COR.

^{2}for the Pooled OLS models are significantly higher in the case of high-income countries than in the case of low-income countries, both for COR and SE).

## 5. Robustness Checks

^{2}= 0.3618) is also validated by the high-income countries’ subsample (Adj R

^{2}= 0.2168), while for the low-income countries’ subsample a weak explicative power is obtained through the Pooled OLS method (Adj R

^{2}= 0.1046). The optimum estimation technique is that of a random-effects model (GLS) and the coefficients of COR are still negative and significant for the worldwide sample and the two subsamples respectively.

## 6. Conclusions

## Author Contributions

## Funding

## Conflicts of Interest

## Appendix A. Description of Subsamples of Countries

Developed Countries (High-Income Countries) (54) | High Income (54) | Australia, Brunei New Zealand, Singapore, South Korea, Austria, Czech Republic, Denmark, Ireland, Italy, Israel, Qatar, Latvia, Lithuania, Netherlands, Norway, Poland, Portugal, Russia, Slovakia, Slovenia, Spain, Sweden, Switzerland, UK, Bahamas, Puerto Rico, Trinidad and Tobago, Uruguay Kuwait, Malta, Saudi Arabia, United Arab Emirates, USA, Hong Kong, Japan, Belgium, Croatia, Cyprus, Estonia, Finland, France, Germany, Greece, Iceland, Luxembourg, Barbados, Chile, Bahrain, Oman, Canada, Macao, Taiwan, Equatorial Guinea |

Developing Countries (Low-Income Countries) (131) | Upper Middle Income (50) | Albania, Algeria, Angola, Argentina, Azerbaijan, Belarus, Belize, Bosnia and Herzegovina, Botswana, Brazil, Bulgaria, Colombia, Costa Rica, Cuba, Dominica, Dominican Republic, Ecuador, Gabon, Grenada, Hungary, Iran, Iraq, Jamaica, Jordan, Kazakhstan, Lebanon, Libya, Macedonia, Malaysia Maldives, Mauritius, Mexico, Montenegro, Namibia, Panama, Peru, Romania, Saint Lucia, Saint Vincent and the Grenadines, Serbia, Seychelles, South Africa, Suriname, Thailand, Tonga, Tunisia, Turkey, Turkmenistan, Venezuela. |

Lower Middle Income (47) | Armenia, Bhutan, Bolivia, Cambodia, Cameroon, Cape Verde, Congo Republic, Côte d´Ivoire, Djibouti, Egypt, El Salvador, Georgia, Ghana, Guatemala, Guyana, Honduras, India, Indonesia, Kiribati, Kosovo, Kyrgyzstan, Laos, Lesotho, Mauritania, Moldova, Mongolia, Morocco, Nicaragua, Nigeria, Pakistan, Papua New Guinea, Paraguay, Philippines, Samoa, Sao Tome and Principe, Senegal, Sri Lanka, Sudan, Swaziland, Syria, Timor-Leste, Ukraine, Uzbekistan, Vanuatu, Vietnam, Yemen, Zambia. | |

Low Income (34) | Afghanistan, Bangladesh, Benin, Burkina Faso, Burundi, Central African Republic, Chad, Comoros, Congo Democratic Republic, Eritrea, Ethiopia, Haiti, Kenya, Korea (North), Gambia, Guinea, Guinea-Bissau, Liberia, Madagascar, Malawi, Mali, Mozambique, Myanmar, Nepal, Niger, Rwanda, Sierra Leone, Somalia, South Sudan, Tajikistan, Tanzania, Togo, Uganda, Zimbabwe. |

## Appendix B. Main Results

**Table A2.**(

**a**) The Estimation of Economic and Sustainable Development measured as LogGDP. Robust (HAC) standard errors, simple regressions, the entire sample. (

**b**) The Estimation of Economic and Sustainable Development measured as LogGDP. Robust (HAC) standard errors, simple regressions, high-income countries’ subsample and low-income countries’ subsample.

(b) | ||||||||||||

LogGDP | (1) | (2) | ||||||||||

Pooled OLS | Fixed-Effects | Random-Effects (GLS) | Pooled OLS | Fixed-Effects | Random-Effects (GLS) | |||||||

Const | 10.4578*** | 8.2927*** | 8.5432*** | 11.0640*** | 10.0848*** | 10.1502*** | ||||||

Corruption | −0.0232*** | 0.0023* | −0.0009 | |||||||||

Shadow economy | −0.0887*** | −0.0547*** | −0.0573*** | |||||||||

Dummy_2008 | 0.0654*** | |||||||||||

Dummy_2009 | −0.0721*** | −0.0596*** | 0.0981*** | 0.0316** | 0.0359*** | |||||||

Dummy_2010 | 0.0238*** | 0.0298*** | 0.0805*** | 0.0605*** | 0.0618*** | |||||||

R^{2} = 0.5709 | LSDV R-squared = 0.9752 | ‘Between’ variance = 0.8885 | R^{2} = 0.4782 | LSDV R-squared = 0.9836 | ‘Between’ variance = 1.2901 | |||||||

Adj R^{2} = 0.5704 | Within R-squared = 0.0181 | ‘Within’ variance = 0.0652 | Adj R^{2} = 0.4773 | Within R-squared = 0.3002 | ‘Within’ variance = 0.0456 | |||||||

No obs. | 1811 | 1665 | ||||||||||

Decision | H = 290.596 with p-value = prob(chi-square(2) > 290.596) = 7.90651 × 10^{−64}, in favour of FEM | H = 22.3252 with p-value = prob(chi-square(3) > 22.3252) = 5.58162 × 10^{−5}, in favour of FEM | ||||||||||

(b) | ||||||||||||

LogGDP | High-Income Countries | Low-Income Countries | ||||||||||

(1) | (2) | (1) | (2) | |||||||||

Pooled OLS | Fixed-Effects | Random-Effects (GLS) | Pooled OLS | Fixed-Effects | Random-Effects (GLS) | Pooled OLS | Fixed-Effects | Random-Effects (GLS) | Pooled OLS | Fixed-Effects | Random-Effects (GLS) | |

Const | 10.6767*** | 10.2339*** | 10.3476*** | 11.1038*** | 11.1206*** | 11.1182*** | 9.0372*** | 7.42*** | 7.5880*** | 8.9061*** | 9.5517*** | 9.4946*** |

Corruption | −0.0111*** | 0.0022 | −0.0011 | −0.0126*** | 0.0024* | 0.0009 | ||||||

Shadow economy | −0.0444*** | −0.0453*** | −0.0452*** | −0.0375*** | −0.0565*** | −0.0551** | ||||||

Dummy_2008 | 0.0960*** | 0.0845*** | 0.0862*** | 0.0562*** | 0.0553*** | 0.0554*** | −0.0438*** | −0.0363** | ||||

Dummy_2009 | −0.0372** | −0.0607*** | −0.0577*** | −0.0787*** | −0.0703*** | 0.0505*** | 0.048*** | |||||

Dummy_2010 | 0.0338*** | 0.0382*** | 0.081*** | 0.0805*** | ||||||||

R^{2} = 0.3144 | LSDV R-squared = 0.9300 | ‘Between’ variance = 0.2227 | R^{2} = 0.3276 | LSDV R-squared = 0.9453 | ‘Between’ variance = 0.23 | R^{2} = 0.2117 | LSDV R-squared = 0.9348 | ‘Between’ variance = 0.7824 | R^{2} = 0.1296 | LSDV R-squared= 0.9557 | ‘Between’ variance = 0.9764 | |

Adj R^{2} = 0.3107 | Within R-squared = 0.0433 | ‘Within’ variance = 0.0281 | Adj R^{2} = 0.3251 | Within R-squared = 0.2698 | ‘Within’ variance = 0.0217 | Adj R^{2} = 0.2111 | Within R-squared = 0.0206 | ‘Within’ variance = 0.0815 | Adj R^{2} = 0.1288 | Within R-squared = 0.3106 | ‘Within’ variance = 0.057 | |

No obs. | 558 | 550 | 1253 | 1115 | ||||||||

Decision | H = 29.9069 with p-value = prob(chi-square(3) > 29.9069) = × 10^{−6} in favour of FEM | H = 0.0103508 with p-value = prob(chi-square(1) > 0.0103508) = 0.918964 in favour of REM | H = 59.4817 with p-value = prob(chi-square(1) > 59.4817) = 1.2344 × 10^{−14}, in favour of FEM | H = 3.65515 with p-value = prob(chi-square(1) > 3.65515) = 0.0558961 in favour of REM |

**Table A3.**(

**a**) The Estimation of Economic and Sustainable Development, measured as Human development index (HDI).Robust (HAC) standard errors, simple regressions, the entire sample. (

**b**)The Estimation of Economic and Sustainable Development, measured as Human development index (HDI). Robust (HAC) standard errors, simple regressions, high-income countries’ subsample and low-income countries’ subsample.

(a) | ||||||||||||

HDI | (1) | (2) | ||||||||||

Pooled OLS | Fixed-Effects | Random-Effects (GLS) | Pooled OLS | Fixed-Effects | Random-Effects (GLS) | |||||||

Const | 0.8833*** | 0.6796*** | 0.6871*** | 0.9381*** | 0.8019 *** | 0.8065*** | ||||||

Corruption | −0.0023*** | 0.00002 | −0.0001 | |||||||||

Shadow economy | −0.0088*** | −0.0041*** | −0.0043*** | |||||||||

Dummy_2008 | −0.0097*** | −0.0093*** | −0.0092*** | −0.0101*** | −0.0101*** | |||||||

Dummy_2009 | −0.0059*** | −0.0055*** | 0.0085*** | |||||||||

Dummy_2010 | 0.0041* | 0.0035** | ||||||||||

R^{2} = 0.5351 | LSDV R-squared = 0.9877 | ‘Between’ variance = 0.0116 | R^{2} = 0.4277 | LSDV R-squared = 0.9914 | ‘Between’ variance = 0.0157 | |||||||

Adj R^{2} = 0.5346 | Within R-squared= 0.0298 | ‘Within’ variance = 0.0003 | Adj R^{2} = 0.4263 | Within R-squared = 0.322 | ‘Within’ variance = 0.0002 | |||||||

No obs. | 1816 | 1700 | ||||||||||

Decision | H = 207.907 with p-value = prob(chi-square(2) > 207.907) = 7.13725 × 10^{−46} in favour of FEM | H = 34.1474 with p-value = prob(chi-square(2) > 34.1474) = 3.84583 × 10^{−8} in favour of FEM | ||||||||||

(b) | ||||||||||||

High-Income Countries | Low-Income Countries | |||||||||||

HDI | (1) | (2) | (1) | (2) | ||||||||

Pooled OLS | Fixed-Effects | Random-Effects (GLS) | Pooled OLS | Fixed-Effects | Random-Effects (GLS) | Pooled OLS | Fixed-Effects | Random-Effects (GLS) | Pooled OLS | Fixed-Effects | Random-effects (GLS) | |

Const | 0.9049 *** | 0.8559*** | 0.8669*** | 0.9444*** | 0.9043*** | 0.9083*** | 0.7691*** | 0.6038*** | 0.609*** | 0.7437*** | 0.75*** | 0.7496*** |

Corruption | −0.0014*** | 0.00002 | −0.0003 | −0.0015*** | 0.00002 | −0.00001 | ||||||

Shadow economy | −0.0049*** | −0.0027*** | −0.0029*** | −0.0042*** | −0.0044*** | −0.0044*** | ||||||

Dummy_2008 | −0.0053*** | −0.0061*** | −0.006*** | −0.0099*** | −0.0081*** | −0.0083*** | −0.0113*** | −0.0111*** | −0.0101*** | −0.0104*** | −0.0104*** | |

Dummy_2009 | −0.0042*** | −0.0058*** | −0.0055*** | −0.0025*** | −0.0023*** | −0.0061*** | −0.0059*** | |||||

Dummy_2010 | −0.0016** | −0.0016 ** | ||||||||||

R^{2} = 0.5849 | LSDV R-squared = 0.9586 | ‘Between’ variance = 0.0013 | R^{2} = 0.4068 | LSDV R-squared = 0.9672 | ‘Between’ variance = 0.002 | R^{2} = 0.196 | LSDV R-squared = 0.9769 | ‘Between’ variance = 0.0133 | R^{2} = 0.1076 | LSDV R-squared = 0.9837 | ‘Between’ variance = 0.016 | |

Adj R^{2} = 0.5826 | Within R-squared = 0.0358 | ‘Within’ variance = 0.0001 | Adj R^{2} = 0.4046 | Within R-squared = 0.1894 | ‘Within’ variance = 0.0001 | Adj R^{2} = 0.1954 | Within R-squared = 0.0309 | ‘Within’ variance = 0.0004 | Adj R^{2} = 0.106 | Within R-squared = 0.351 | ‘Within’ variance = 0.0003 | |

No obs. | 545 | 550 | 1271 | 1150 | ||||||||

Decision | H = 68.6927 with p-value = prob(chi-square(3) > 68.6927) = 8.13129 × 10^{−15} in favour of FEM | H = 6.50989 with p-value = prob(chi-square(1) > 6.50989) = 0.0107276 in favour of FEM | H = 35.9574 with p-value = prob(chi-square(1) > 35.9574) = 2.01683 × 10^{−9}, in favour of FEM | H = 0.242071 with p-value = prob(chi-square(2) > 0.242071) = 0.886002 in favour of REM |

**Table A4.**(

**a**) The Estimation of Economic and Sustainable Development, measured as Environmental performance index (EPI). Robust (HAC) standard errors, simple regressions, the entire sample. (

**b**) The Estimation of Economic and Sustainable Development, measured as Environmental performance index (EPI). Robust (HAC) standard errors, simple regressions, high-income countries’ subsample and low-income countries’ subsample.

(a) | ||||||||||||

EPI | (1) | (2) | ||||||||||

Pooled OLS | Fixed-Effects | Random-Effects (GLS) | Pooled OLS | Fixed-Effects | Random-Effects (GLS) | |||||||

Const | 69.2211*** | 49.3478*** | 59.5931*** | 73.9634*** | 61.893*** | 68.8388 *** | ||||||

Corruption | −0.2182*** | 0.0131 | −0.1087*** | |||||||||

Shadow economy | −0.7926*** | −0.381** | −0.622** | |||||||||

Dummy_2008 | 0.6525* | |||||||||||

Dummy_2009 | 0.9577** | 0.5285* | 1.4381*** | 0.9039*** | 1.215*** | |||||||

Dummy_2010 | 0.8111** | 0.6151** | 0.8344*** | 0.7651*** | 0.804*** | |||||||

R^{2} = 0.4364 | LSDV R-squared = 0.8159 | ‘Between’ variance = 115.956 | R^{2} = 0.3423 | LSDV R-squared = 0.8194 | ‘Between’ variance = 126.982 | |||||||

Adj R^{2} = 0.4349 | Within R-squared = 0.0004 | ‘Within’ variance = 58.2328 | Adj R^{2} = 0.3409 | Within R-squared = 0.016 | ‘Within’ variance = 57.245 | |||||||

No obs. | 1474 | 1381 | ||||||||||

Decision | H = 85.5988 with p-value = prob(chi-square(4) > 85.5988) = 1.13223 × 10^{−17} in favour of FEM | H = 14.8036 with p-value = prob(chi-square(2) > 14.8036) = 0.000610163 in favour of FEM | ||||||||||

(b) | ||||||||||||

High-Income Countries | Low-Income Countries | |||||||||||

EPI | (1) | (2) | (1) | (2) | ||||||||

Pooled OLS | Fixed-Effects | Random-Effects (GLS) | Pooled OLS | Fixed-Effects | Random-Effects (GLS) | Pooled OLS | Fixed-Effects | Random-Effects (GLS) | Pooled OLS | Fixed-Effects | Random-Effects (GLS) | |

Const | 71.6847*** | 64.9243*** | 69.8618*** | 79.2721*** | 55.7528*** | 73.9757*** | 57.1787*** | 41.7543*** | 47.1726*** | 52.9833*** | 63.3299*** | 58.5152*** |

Corruption | −0.1875*** | 0.0166 | −0.1494*** | −0.122*** | 0.0189 | −0.0278* | ||||||

Shadow economy | −0.7544*** | 0.5387** | −0.4697*** | −0.2736*** | −0.5745*** | −0.4383*** | ||||||

Dummy_2008 | 3.652*** | 3.183*** | 3.4113*** | 2.6765*** | 3.7044*** | 2.8878*** | −0.9128** | −1.423*** | −1.2449*** | −1.1898*** | −1.0997*** | −1.06*** |

Dummy_2009 | 3.9266*** | 3.3761*** | 3.665*** | 4.3123*** | 2.9273*** | 3.9924*** | −0.7312* | −1.2768*** | −1.0872*** | −0.633* | ||

Dummy_2010 | 4.0888*** | 3.722*** | 3.8615*** | 4.1734*** | 3.5953*** | 4.0311*** | −0.8037** | −0.9408*** | −0.8726*** | −0.7364** | ||

R^{2} = 0.2248 | LSDV R-squared = 0.6216 | ‘Between’ variance = 53.3006 | R^{2} = 0.2326 | LSDV R-squared = 0.6012 | ‘Between’ variance = 46.109 | R^{2} = 0.1348 | LSDV R-squared = 0.7253 | ‘Between’ variance = 107.561 | R^{2} = 0.0523 | LSDV R-squared = 0.736 | ‘Between’ variance = 103.936 | |

Adj R^{2} = 0.2178 | Within R-squared = 0.0406 | ‘Within’ variance = 70.7438 | Adj R^{2} = 0.2257 | Within R-squared = 0.053 | ‘Within’ variance = 72.736 | Adj R^{2} = 0.1314 | Within R-squared = 0.0079 | ‘Within’ variance = 51.4123 | Adj R^{2} = 0.0482 | Within R-squared = 0.0514 | ‘Within’ variance = 46.8788 | |

No obs. | 449 | 451 | 1025 | 930 | ||||||||

Decision | H = 24.992 with p-value = prob(chi-square(3) > 24.992) = 1.55002 × 10^{−5} in favour of FEM | H = 27.7698 with p-value = prob(chi-square(2) > 27.7698) = 9.32974 × 10^{−7} in favour of FEM | H = 23.3646 with p-value = prob(chi-square(4) > 23.3646) = 0.000107065, in favour of FEM | H = 5.76895 with p-value = prob(chi-square(2) > 5.76895) = 0.0558843, in favour of REM |

## Appendix C. Robustness Checks

**Table A5.**Robustness checks for GSCI as a function of Corruption (COR). Robust (HAC) standard errors.

GSCI | The Entire Sample | High-Income Countries | Low-Income Countries | ||||||
---|---|---|---|---|---|---|---|---|---|

Pooled OLS | Fixed-Effects | Random-Effects (GLS) | Pooled OLS | Fixed-effects | Random-Effects (GLS) | Pooled OLS | Fixed-Effects | Random-Effects (GLS) | |

Const | 48.9376*** | 47.0725*** | 47.728*** | 51.3328*** | 50.2577*** | 50.3941*** | 44.1331 *** | 45.7811*** | 44.2407*** |

Corruption | −0.0767*** | −0.0553*** | −0.0669*** | −0.1075*** | −0.07553*** | −0.0962*** | −0.0384*** | −0.0533*** | −0.0413*** |

R^{2} = 0.3625 | LSDV R-squared = 0.8828 | ‘Between’ variance = 23.2837 | R^{2} = 0.22 | LSDV R-squared = 0.8632 | ‘Between’ variance = 36.8468 | R^{2} = 0.1062 | LSDV R-squared = 0.7966 | ‘Between’ variance = 14.1905 | |

Adj R^{2} = 0.3618 | Within R-squared = 0.0337 | ‘Within’ variance = 6.1245 | Adj R^{2} = 0.2168 | Within R-squared = 0.0141 | ‘Within’ variance = 8.0074 | Adj R^{2} = 0.1046 | Within R-squared = 0.0468 | ‘Within’ variance = 5.31959 | |

No obs. | 817 | 246 | 571 | ||||||

Decision | H = 1.4165 with p-value = prob(chi-square(1) > 1.4165) = 0.23398 in favour of REM | H = 0.293443 with p-value = prob(chi-square(1) > 0.293443) = 0.588022 in favour of REM | H = 1.85992 with p-value = prob(chi-square(1) > 1.85992) = 0.172634, in favour of REM |

**Table A6.**Robustness checks for GSCI as a function of Shadow economy (SE), Robust (HAC) standard errors.

GSCI | The Entire Sample | HI Countries’ Subsample | LI Countries’ Subsample | ||||||
---|---|---|---|---|---|---|---|---|---|

Pooled OLS | Fixed-Effects | Random-Effects (GLS) | Pooled OLS | Fixed-Effects | Random-Effects (GLS) | Pooled OLS | Fixed-Effects | Random-Effects (GLS) | |

Const | 50.9403*** | 46.2363*** | 49.8516 *** | 55.5348*** | 36.5824*** | 52.4635*** | 43.3128 *** | 47.3046 *** | 44.4019*** |

Shadow economy | −0.3001*** | −0.1316** | −0.2618*** | −0.4478*** | 0.6383*** | −0.2751*** | −0.0986*** | −0.2195*** | −0.1315*** |

R^{2} = 0.3076 | LSDV R-squared= 0.8725 | ‘Between’ variance = 21.9352 | R^{2} = 0.2782 | LSDV R-squared = 0.856 | ‘Between’ variance = 25.2566 | R^{2} = 0.0481 | LSDV R-squared = 0.7781 | ‘Between’ variance = 13.3079 | |

Adj R^{2} = 0.3067 | Within R-squared= 0.0076 | ‘Within’ variance = 6.5023 | Adj R^{2} = 0.2752 | Within R-squared = 0.0438 | ‘Within’ variance = 8.0307 | Adj R^{2} = 0.0462 | Within R-squared = 0.0325 | ‘Within’ variance = 5.48942 | |

No obs. | 763 | 249 | 514 | ||||||

Decision | H = 5.9242 with p-value = prob(chi-square(1) > 5.9242) = 0.0149343 in favour of FEM | H = 22.5328 with p-value = prob(chi-square(1) > 22.5328) = 2.06589 × 10^{−6} in favour of FEM | H = 3.16691 with p-value = prob(chi-square(1) > 3.16691) = 0.0751446, in favour of REM |

## Appendix D. X-Y Scatter Plots

COR on the X-Axis | SE on the X-Axis | |

GDP on the Y-axis | ||

HDI on the Y-axis | ||

EPI on the Y-axis | ||

Source: Authors’ processings. |

## Appendix E. Graphs for the Actual (+) and Fitted (×) Values of the GDP, HDI and EPI Respectively, through the Optimal Estimation Technique, Using COR and SE As Exogenous Variables

COR as Independent Variable | SE as Independent Variable | |

Model (1) FEM | Model (2) FEM | |

The Estimation of GDP, the entire sample, from Table A2a | ||

Model (1) FEM | Model (2) FEM | |

The Estimation of HDI, the entire sample, from Table A3a | ||

Model (1) FEM | Model (2) FEM | |

The Estimation of EPI, the entire sample, from Table A4a | ||

Source: Authors’ processings. |

## References

- Mauro, P. Corruption and growth. Q. J. Econ.
**1995**, 110, 681–712. [Google Scholar] [CrossRef] - Djankov, S.; La Porta, R.; Lopez-de-Silanes, F.; Shleifer, A. The regulation of entry. Q. J. Econ.
**2002**, 117, 1–37. [Google Scholar] [CrossRef] - Dreher, A.; Schneider, F. Corruption and the Shadow Economy: An Empirical Analysis; CREMA Working Paper Series 2006-01; Center for Research in Economics, Management and the Arts (CREMA): Zürich, Switzerland, 2006. [Google Scholar]
- Sahakyan, N.; Stiegert, K.W. Corruption and firm performance. East. Eur. Econ.
**2012**, 50, 5–27. [Google Scholar] [CrossRef] - Groşanu, A.; Boţa-Avram, C.; Răchişan, P.R.; Vesselinov, R.; Tiron-Tudor, A. The influence of country-level governance on business environment and entrepreneurship: A global perspective. Amfiteatru Econ. J.
**2015**, 17, 62–78. [Google Scholar] - Beck, P.; Mahler, M. A comparison of bribery and bidding in thin markets. Econ. Lett.
**1986**, 20, 1–5. [Google Scholar] [CrossRef] [Green Version] - Caselli, F.; Michaels, G. Do oil windfalls improve living standards? Evidence from Brazil. Am. Econ. J. Appl. Econ.
**2013**, 5, 208–238. [Google Scholar] [CrossRef] [Green Version] - Jiang, T.; Nie, H. The stained China miracle: Corruption, regulation, and firm performance. Econ. Lett.
**2014**, 123, 366–369. [Google Scholar] [CrossRef] - Rodrik, D.; Subramanian, A.; Trebbi, F. Institutions rule: The primacy of institutions over geography and integration in economic development. J. Econ. Growth
**2004**, 9, 131–165. [Google Scholar] [CrossRef] - Stojanovi´c, I.; Ateljevi´c, J.; Stevi´c, R.S. Good Governance as a Tool of Sustainable Development. Eur. J. Sustain. Dev.
**2016**, 5, 558–573. [Google Scholar] [CrossRef] - Zaman, G.; Goschin, Z. Shadow economy and economic growth in Romania. Cons and pros. Procedia Econ. Financ.
**2015**, 22, 80–87. [Google Scholar] [CrossRef] [Green Version] - World Bank. Anticorruption; World Bank: Washington, DC, USA, 2009. [Google Scholar]
- Paldam, M. The big pattern of corruption: Economics, culture and the see saw dynamics. Eur. J. Political Econ.
**2002**, 18, 215–240. [Google Scholar] [CrossRef] - Achim, M.; Borlea, N.S. Determinants of the European funds absorption 2007–2013 in European Union Member States. In Proceedings of the WEI International Academic Conference in Vienna, Vienna, Austria, 12–15 April 2015; pp. 174–188. [Google Scholar]
- Fjeldstad, O.H. Tax Evasion and Corruption in Local Governments in Tanzania: Alternative Economic Approaches; Working Paper 14; Chr. Michelsen Institute: Bergen, Norway, 1996; pp. 1–31. [Google Scholar]
- Fjeldstad, O.H. Fighting fiscal corruption: Lessons from the Tanzania revenue authority. In Public Administration and Development; Wiley: Hoboken, NJ, USA, 2003; Volume 23, pp. 165–175. [Google Scholar]
- Kaufman, D. Can Corruption Adversely Affect Public Finance in Industrialized Countries; Brookings Institution Opinions; Brookings Institution: Washington, DC, USA, 19 April 2010. [Google Scholar]
- Ivanyna, M.; Moumouras, A.; Rangazas, P. The Culture of Corruption, Tax Evasion, and Optimal Tax Policy. 2010. Available online: http://www2.warwick.ac.uk/fac/soc/economics/events/seminars-workshopsconferences/conferences/mysore/programme/ivanyna.pdf (accessed on 15 March 2019).
- Bordean, O.N. Business students’ perceptions on corporate governance. Int. J. Arts Commer.
**2012**, 1, 173–184. [Google Scholar] - De Rosa, D.; Gooroochurn, N.; Gorg, H. Corruption and Productivity: Firm-Level. Evidence from the BEEPS Survey; Kiel Working Paper No. 1632; World Bank: Washington, DC, USA, 2010. [Google Scholar]
- Achim, M.V. Corruption, income and business development. J. Int. Bus. Entrep. Dev.
**2017**, 10, 85–100. [Google Scholar] [CrossRef] - Absalyamova, S.; Absalyamov, T.; Khusnullova, A.; Mukhametgalieva, C. The impact of corruption on the sustainable development of human capital. J. Phys. Conf. Ser.
**2016**, 738, 1–6. [Google Scholar] [CrossRef] [Green Version] - Forson, J.A.; Buracom, P.; Chen, G.; Baah-Ennumh, T.Y. Genuine Wealth Per Capita as a Measure of Sustainability and the Negative Impact of Corruption on Sustainable Growth in Sub-Sahara Africa. S. Afr. J. Econ.
**2017**, 85, 178–195. [Google Scholar] [CrossRef] - Kirchler, E. The Economic Psychology of Tax Behavior; Cambridge University Press: Cambridge, UK, 2007. [Google Scholar]
- Borlea, S.; Achim, M.V.; Miron, M. Corruption, Shadow Economy and Economic Growth: An Empirical Survey Across the European Union Countries. Studia Univ. Vasile Goldis Arad Ser.
**2017**, 27, 19–32. [Google Scholar] [CrossRef] [Green Version] - Hamra, W. Bribery in international business transactions and the OECD convention: Benefits and limitations. Bus. Econ.
**2000**, 35, 33–46. [Google Scholar] - Ruzek, W. The Informal Economy as a Catalyst for Sustainability. Sustainability
**2015**, 7, 23–34. [Google Scholar] [CrossRef] [Green Version] - Husted, B.W. Wealth, culture, and corruption. J. Int. Bus. Stud.
**1999**, 30, 339–359. [Google Scholar] [CrossRef] - Treisman, D. The causes of corruption: A cross-national study. J. Public Econ.
**2000**, 76, 399–457. [Google Scholar] [CrossRef] [Green Version] - Gundlach, E.; Paldam, M. The transition of corruption: From poverty to honesty. Econ. Lett.
**2009**, 103, 146–148. [Google Scholar] [CrossRef] [Green Version] - Transparency International. Corruption Perception Index. Available online: https://www.transparency.org/ (accessed on 20 February 2019).
- Schneider, F.; Klinglmair, R. Shadow Economies Around the World: What Do We Know? IZA Discussion Paper No. 1043. CESifo Working Paper Series No. 1167. 2004. Available online: https://ssrn.com/abstract=518526 (accessed on 20 August 2019).
- Orviska, M.; Hudson, J. Tax evasion, civic duty and the law abiding citizen. Eur. J. Political Econ.
**2003**, 19, 83–102. [Google Scholar] [CrossRef] - World Bank. Corruption, helping countries combat: The Role of the World Bank. In Poverty Reductionand Economic Management; World Bank: Washington, DC, USA, 1997; p. 1. [Google Scholar]
- Medina, L.; Schneider, F. Shadow Economies Around the World: What Did We Learn Over the Last 20 Years; WP/18/17; International Monetary Fund Working Paper: Washington, DC, USA, 2018. [Google Scholar]
- Schneider, F.; Buehn, A. Shadow Economy: Estimation Methods, Problems, Results and Open Questions. Open Econ.
**2018**, 1, 1–29. [Google Scholar] [CrossRef] [Green Version] - Schneider, F.; Buehn, A.; Montenegro, C.E. New Estimates for the Shadow Economies All over the World. Int. Econ. J.
**2010**, 24, 443–461. [Google Scholar] [CrossRef] - Deardorff, A.V. Deardorffs’ Glossary of International Economics. 2016. Available online: http://www-personal.umich.edu/~alandear/glossary/e.html (accessed on 21 December 2019).
- Torgler, B.; Schneider, F. The Impact of Tax Morale and Institutional Quality on the Shadow Economy. J. Econ. Psychol.
**2009**, 30, 228–245. [Google Scholar] [CrossRef] [Green Version] - Boța-Avram, C.; Groșanu, A.; Răchișan, P.R.; Gavriletea, M.D. The Bidirectional Causality between Country-Level Governance, Economic Growth and Sustainable Development: A Cross-Country Data Analysis. Sustainability
**2018**, 10, 502. [Google Scholar] [CrossRef] [Green Version] - World Bank. World Development Indicators; The World Bank: Washington, DC, USA, 2019; Available online: https://openknowledge.worldbank.org (accessed on 10 August 2019).
- Stiglitz, J.E.; Sen, A.; Fitoussi, J.P. Mismeasuring Our Lives: Why GDP Doesn’t Add Up, Commission on the Measurement of Economic Performance and Social Progress; New Press: New York, NY, USA, 2010. [Google Scholar]
- Aidt, T.S. Corruption and Sustainable Development; No. CWPE 1061. Cambridge, UK, 2010. Available online: https://www.repository.cam.ac.uk/bitstream/handle/1810/242086/cwpe1061.pdf;jsessionid533A12327621029A0EF0B686DD27C5E5C?sequence51 (accessed on 12 September 2019).
- World Commission on Environment and Development. Our Common Future; Oxford University Press: Oxford, UK, 1987; p. 27. ISBN 019282080X. [Google Scholar]
- UNDP. Human Development Reports. Available online: http://hdr.undp.org (accessed on 10 September 2019).
- Murshed, M.; Mredula, F.A. Impacts of Corruption on Sustainable Development: A Simultaneous Equations Model Estimation Approach. J. Account. Financ. Econ.
**2018**, 8, 109–133. [Google Scholar] - Yale University. Environmental Performance Index. Available online: https://epi.envirocenter.yale.edu (accessed on 15 October 2019).
- Esty, D.; Levy, M.A.; Kim, C.H.; de Sherbinin, A.; Srebotnjak, T.; Mara, V. Environmental Performance Index; Yale Center for Environmental Law and Policy: New Haven, CT, USA, 2008. [Google Scholar]
- Emerson, J.W.; Hsu, A.; Levy, M.A.; de Sherbinin, A.; Mara, V.; Esty, D.C.; Jaiteh, M. Environmental Performance Index and Pilot Trend Environmental Performance Index; Yale Center for Environmental Law and Policy: New Haven, CT, USA, 2012; Available online: http://epi.yale.edu (accessed on 10 May 2012).
- Gallego-Alvarez, I.; Vicente-Galindo, M.; Galindo-Villardón, M.; Rodríguez-Rosa, M. Environmental Performance in Countries Worldwide: Determinant Factors and Multivariate Analysis. Sustainability
**2014**, 6, 7807–7832. [Google Scholar] [CrossRef] [Green Version] - World Bank Group. World Bank report ‘Country and Lending Groups’ Report. 2015. Available online: https://www.worldbank.org/ (accessed on 10 February 2019).
- Williams, C.C. Towards a Commodified World? Re-Reading Economic Development in Western Economies. Int. J. Econ. Dev.
**2004**, 6, 1–21. [Google Scholar] [CrossRef] [Green Version] - Achim, M.V.; Borlea, N.S. Economic and Financial Crime. Corruption, Shadow Economy and Money Laundering. Theoretical and Practical Approaches; Economica Publishing House: Bucharest, Romania, 2019. [Google Scholar]
- SolAbility Sustainable Intelligence. The Global Competitiveness Reports. Available online: http://solability.com/the-global-sustainable-competitiveness-index (accessed on 25 October 2019).
- Khan, M.H. Governance, Economic Growth and Development Since the 1960s; UN DESA Working Paper No. 54; DESA: New York, NY, USA, 2007. [Google Scholar]

**Table 1.**Summary Statistics for the entire sample and the subsamples of high-income (HI) and low-income (LI) countries respectively, using the observations 1:01–185:11 (missing values were skipped).

Variable | Mean | Median | S.D. | Min | Max |
---|---|---|---|---|---|

GDPpercap_ALL | 1.31 × 10^{4} | 4.70 × 10^{3} | 1.89 × 10^{4} | 151. | 1.19 × 10^{5} |

GDPpercap_HI | 3.57 × 10^{4} | 3.18 × 10^{4} | 2.11 × 10^{4} | 5.22 × 10^{3} | 1.19 × 10^{5} |

GDPpercap_LI | 3.54 × 10^{3} | 2.60 × 10^{3} | 3.22 × 10^{3} | 151. | 1.57 × 10^{4} |

HDI_ALL | 0.677 | 0.706 | 0.161 | 0.286 | 0.949 |

HDI_HI | 0.853 | 0.865 | 0.0604 | 0.569 | 0.949 |

HDI_LI | 0.605 | 0.635 | 0.132 | 0.286 | 0.836 |

EPI_ALL | 50.12 | 49.67 | 16.63 | 14.68 | 88.79 |

EPI_HI | 66.24 | 67.97 | 13.00 | 21.57 | 88.79 |

EPI_LI | 43.37 | 43.75 | 12.94 | 14.68 | 87.67 |

COR_ALL | 86.0 | 84.5 | 50.4 | 1.00 | 182. |

COR_HI | 33.3 | 26.0 | 29.9 | 1.00 | 172. |

COR_LI | 109. | 110. | 38.9 | 11.0 | 182. |

SE_ALL | 29.1 | 29.4 | 12.3 | 6.16 | 69.1 |

SE_HI | 18.1 | 16.6 | 7.81 | 6.16 | 39.9 |

SE_LI | 34.5 | 33.7 | 10.3 | 11.7 | 69.1 |

LogGDP | HDI | EPI | Corr | SE | |
---|---|---|---|---|---|

LogGDP | 1.0000 | ||||

HDI | 0.9220 | 1.0000 | |||

EPI | 0.7726 | 0.8027 | 1.0000 | ||

COR | −0.7539 | −0.7295 | −0.6625 | 1.0000 | |

SE | −0.6900 | −0.6502 | −0.5975 | 0.6548 | 1.0000 |

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Hoinaru, R.; Buda, D.; Borlea, S.N.; Văidean, V.L.; Achim, M.V.
The Impact of Corruption and Shadow Economy on the Economic and Sustainable Development. Do They “Sand the Wheels” or “Grease the Wheels”? *Sustainability* **2020**, *12*, 481.
https://doi.org/10.3390/su12020481

**AMA Style**

Hoinaru R, Buda D, Borlea SN, Văidean VL, Achim MV.
The Impact of Corruption and Shadow Economy on the Economic and Sustainable Development. Do They “Sand the Wheels” or “Grease the Wheels”? *Sustainability*. 2020; 12(2):481.
https://doi.org/10.3390/su12020481

**Chicago/Turabian Style**

Hoinaru, Răzvan, Daniel Buda, Sorin Nicolae Borlea, Viorela Ligia Văidean, and Monica Violeta Achim.
2020. "The Impact of Corruption and Shadow Economy on the Economic and Sustainable Development. Do They “Sand the Wheels” or “Grease the Wheels”?" *Sustainability* 12, no. 2: 481.
https://doi.org/10.3390/su12020481