**5. Consistency and robustness of empirical evidence**

According to Huber (1973), as further evidence of the robustness of results, it is appropriate to apply the robust regression (RREG) estimator to cope with possible outliers from our da‐ taset. These outliers can impair the stability and reliability of results. Such as in Marques and Fuinhas (2012b), robustness is analyzed by providing the robust regression with Huber and Tukey weight functions, as presented in table 6.

As shown from this additional assessment of the robustness of results, the variables main‐ tain their signs, though with small differences in significance levels. In general, the robust regression validates the main results of the estimations.


**Ind. Variables**

**RREG RREG RREG (VII) (VIII) (IX)**

http://dx.doi.org/10.5772/52159

71

Notes: RREG – Robust Regression. The F-test has normal distribution N(0,1) and tests the null hypothesis of non-signifi‐ cance of all estimated parameters. JST - Joint Significance Test. JST is a Wald χ <sup>2</sup> test with the null hypothesis of *H*<sup>0</sup> =β<sup>1</sup> =β<sup>2</sup> =β<sup>3</sup> =β<sup>4</sup> =0, with β1, β2, β3, β<sup>4</sup> representing the coefficient of FINPOLc,t, EDUPOLc,t, COOPPOLc,t and CUTPOLc,t, respectively. LRT - Linear Restriction Test has the null hypothesis of *H*<sup>0</sup> =β<sup>1</sup> + β<sup>2</sup> + β<sup>3</sup> + β<sup>4</sup> =0. Standard errors are report‐ ed in brackets. All estimates were controlled to include time effects, but they are not reported for reasons of simplicity.

On the Public Policies Supporting Renewables and Wind Power Overcapacity: Insights into the European Way Forward

As stated above, renewable energy sources, particularly wind and solar, suffer from the in‐ termittency phenomenon. This phenomenon could cause overcapacity. There are several factors that may influence overcapacity, such as conventional energy sources and renewable energy sources, socio-economic and energy policies. Our results allow us to explore and dis‐

As shown in the models, the results for conventional energy sources show a negative effect of fossil fuels on overcapacity, more specifically coal and gas power plants. With greater use of coal and gas, the effect of wind overcapacity is reduced. Two main reasons can be at the origin of this effect: (i) intermittency leads to the uncertainty of energy generation and the need to ensure a continuous electricity supply. It requires the existence of fossil fuels like coal and gas to backup power. With more dependence on these sources in peak-load peri‐ ods, electricity generation is simultaneously based on renewables and fossil fuels in order to meet electricity demand and this implies a reduction in wind overcapacity; and (ii) in line with Marques et al. (2010) the results for fossil fuels sustain a lobbying effect in the electrici‐ ty generation industry. This effect promotes the growth of fossil fuels to the detriment of re‐ newables due to more stringent energy policies (Fredriksson et al., 2004). The first sites for installation of wind farms are usually the most efficient ones, and, in some countries, the de‐ ployment of renewables is still in its early stages because fossil fuels still have high shares in total electricity generation. Therefore, some countries still benefit from better sites with high wind speeds and from better capacity factor and, as a consequence, wind overcapacity tends

Nevertheless, assuming that fossil fuels could have a positive effect on overcapacity, an in‐ crease in the share of coal, gas and oil would provoke a substitution effect in the electrici‐

cuss them individually and suggest some guidance for energy policy and measures.

N 218 218 218 R2 63.59 0.6920 0.7093 F (N(0,1)) 15.48\*\*\* 19.92\*\*\* 17.93\*\*\*

\*\*\*, \*\*, denote significance at 1 and 5% significance levels respectively.

**Table 6.** Results from Robust Regression – Dependent variable *WOCAPc,t*

**6. Conventional energy sources and backup**

to be lower, since wind power is installed in optimum sites.

On the Public Policies Supporting Renewables and Wind Power Overcapacity: Insights into the European Way Forward http://dx.doi.org/10.5772/52159 71


Notes: RREG – Robust Regression. The F-test has normal distribution N(0,1) and tests the null hypothesis of non-signifi‐ cance of all estimated parameters. JST - Joint Significance Test. JST is a Wald χ <sup>2</sup> test with the null hypothesis of *H*<sup>0</sup> =β<sup>1</sup> =β<sup>2</sup> =β<sup>3</sup> =β<sup>4</sup> =0, with β1, β2, β3, β<sup>4</sup> representing the coefficient of FINPOLc,t, EDUPOLc,t, COOPPOLc,t and CUTPOLc,t, respectively. LRT - Linear Restriction Test has the null hypothesis of *H*<sup>0</sup> =β<sup>1</sup> + β<sup>2</sup> + β<sup>3</sup> + β<sup>4</sup> =0. Standard errors are report‐ ed in brackets. All estimates were controlled to include time effects, but they are not reported for reasons of simplicity. \*\*\*, \*\*, denote significance at 1 and 5% significance levels respectively.

**Table 6.** Results from Robust Regression – Dependent variable *WOCAPc,t*

**Ind. Variables**

70 New Developments in Renewable Energy

COALSHc,t

GASSHc,t

OILSHc,t

CFNUCLc,t

CFHYDc,t

WASTSHc,t

WINDGRc,t

SOLSHc,t

POPDENSc,t

LNGDPPCc,t

ALLPOLc,t

NORMPOLc,t

FISCPOLc,t

FINPOLc,t

EDUPOLc,t

COOPPOLc,t

CUTPOLc,t

CONST

**RREG RREG RREG (VII) (VIII) (IX)**

> -0.0130 (0.0115)




0.0270 (0.0253)

0.3410\*\*\* (0.0740)

0.0003\*\*\* (0.0000)


0.0002\*\*\* (0.0000)


0.0030\*\*\* (0.0004)


1.0238\*\*\* (0.0399)





0.0015 (0.0284)

0.3598\*\*\* (0.0827)

0.0003\*\*\* (0.0000)


0.0002\*\*\* (0.0000)


0.0030\*\*\* (0.0005)




0.0011 (0.0011)

0.0012 (0.0008)

1.0377\*\*\* (0.0457)





0.0084 (0.0270)

0.3199\*\*\* (0.0808)

0.0003\*\*\* (0.0000)

1.6675 (1.3598)

0.0002\*\*\* (0.0000)


0.0007\*\*\* (0.0002)

1.0712\*\*\* (0.0464)
