**Development Financial and economic crime it 0 1** = + + **it it** β β ε (1)

where: Developmentit – proxy for the development dimensions of country i in year t; it includes:

GDPit – the per capita current USD gross domestic product of country i in year t *and.*

HDIit – the human development index of country i in year t;

β0 - intercept;

β1 - linear effect parameter;

Financial and economic crimeit – proxy for the financial and economic crime dimensions of country i in year t (Corruptionit*,* Shadow Economyit*,* Money Launderingit*,* Cybercrimeit);

εit - the prediction error.

#### **3.2 Results and discussions**

The results have been synthesised within **Table 3**. Basically **Table 3** contains the estimations of the GDP economic prosperity proxy as a function of financial and economic crime proxies: the independent variables are, on turn, Corruption in model (1a), Shadow Economy in model (2a), Money Laundering in model (3a) and *Economic and Financial Crimes and the Development of Society DOI: http://dx.doi.org/10.5772/intechopen.96269*

#### **Figure 12.**

*Economic and human development as a function of financial and economic crime proxies. Source: Authors' processings.*

Cybercrime in model (4a). In the same manner, **Table 3** contains the estimations of the HDI as a function of financial and economic crime proxies (models (1b)- (4b)). Except for models (3a) and (3b) which are log-linear models, all the other models are log–log models. As such, models (1a), (2a), (4a), (1b), (2b) and (4b) are


#### **Table 2.**

*Correlation matrix.*

commonly referred to as *elastic* and the coefficients of financial and economic crime proxies are referred to as *elasticities* [38]. Basically we want to estimate the impact held by various financial and economic crimes' proxies upon economic prosperity (**Table 3**) and human development (**Table 3**).

By simply comparing the estimated coefficients from **Table 3**, one may easily notice that the absolute values from **Table 3** are larger than the ones from **Table 3**, thus we somehow expect a more pronounced impact of the vector of financial and economic crime proxies from Eq. 1 upon the economic development than upon human sustainable development.

For log–log models, the interpretations are considered as an expected percentage change in development when the financial and economic crime proxy increases by some percentage. For model (1a) in terms of effects of changes in Corruption on GDP (both unlogged) we have that multiplying Corruption by *e* will multiply the expected value of GDP by e−0.4577. In other words, a 1% increase in Corruption multiplies GDP by e−0.0045, so actually GDP is reduced by 4.6%, everything else unchanged. The effect of Corruption upon human development is estimated through the simple regression modelling from model (1b): a 1% increase in Corruption multiplies HDI by 0.9997, so actually HDI is reduced by 0.03%, everything else unchanged. So, corruption on the one hand and economic and human development on the other are indirectly related, the decrease in corruption having positive effects upon development.

The impact of shadow economy upon development proxies is estimated through models (2a) and (2b). As such, the −1.3301 elasticity from model (2a) (**Table 3**) gives that a 10% increase in Shadow Economy multiplies GDP by 0.8809, so actually we get an 11.91% reduction of economic prosperity. In a similar manner, a 10% increase in Shadow Economy reduces HDI by 7.01%, everything else unchanged, from model (2b) (**Table 3**). As expected, the more shadow economy evolves, the less developmental benefits it brings. All in all, the negative effect of corruption and shadow economy upon economic prosperity and human development is validated, with a stronger impact upon economic development.

Model (4a) estimates the effect of cybercrime upon development. Multiplying Cybercrime by *e* ≈ 2.72 multiplies GDP by e1.3342 = 3.7969, i.e. increases the expected GDP by about 279.69%. Further on, model (4b) estimates the multiplicative changes in both Cybercrime and HDI: multiplying Cybercrime by *e* multiplies HDI by 1.0967, i.e. increases the expected HDI by about 9.67%. The graphical representations and the correlation coefficients depicted a direct relationship between the

#### *Economic and Financial Crimes and the Development of Society DOI: http://dx.doi.org/10.5772/intechopen.96269*


**Table 3.**

*Simple regression modelling.*

evolution of cybercrime and that of development proxies. That direct correlation is validated by out tabled coefficients: cybercrime and both economic and human development move in tandem, like a hand in hand walk. It seems that the more developed the economic conditions are and the more evolved people have become, cybercrime is conferred a boost, especially in the last years.

Nonetheless, the interpretation of the log-linear model (3a) (**Table 3**) is the following: each one unit increase in Money Laundering increases LogGDP by 0.1359. For the untransformed GDP, each one unit increase of Money Laundering increases economic prosperity by a multiple of e0.1359 = 1.1456 or a 14.56% increase. Then, model (3b) provides the following estimation: each one unit increase of Money Laundering increases human development by a multiple of 1.0034, that is a 0.34% increase. Thus, there's a direct relationship between money laundering and development, just like between cybercrime and development. Somehow, money laundering has a positive effect on prosperity, both economic and human, influencing it directly. We have not tested for causality as this was not our research purpose, but money laundering moves just like development does.
