The Influence of Renewable Energy
Production on the Forward Electricity Price
S.J. Welling*
Supervisor: Dr. G.T.J. Zwart
Master’s Thesis Finance
University of Groningen
12 June 2019
Abstract
Renewable energy production changes the way the market prices electricity. The objective of this study is to evaluate the impact of this change on the forward electricity market. This panel study of eight European countries uses data from Bloomberg Finance L.P. and the European Network of Transmission System Oper-ators for Electricity. The effect of the changing share of renewables in the energy mix is estimated on prices and the volatility level in the forward market from 2010 to 2018. The results of this study suggest a negative relationship between renewable energy production and the forward electricity price. In addition, the forward elec-tricity price does not become more volatile due to the changing share of renewable energy production. Therefore, it can be concluded that for investors the expected price of electricity has not become more difficult to hedge against.
Keywords: renewable electricity, forward electricity price JEL-codes: C23, G13, Q41
1
Introduction
In recent years the amount of renewable energy in the Netherlands increased towards
6.6% of the total energy consumption in 2017 (Meurink, Muller, & Segers, 2017). This
is a ten percent increase with respect to the 6.0% in 2016. Even though the share of
renewable energy consumption in The Netherlands has increased over the last years, it
remains uncertain whether the 14 percent quotum set by the European Union for 2020 will
be met (European Union, 2009). Similarly, several European countries like France and
Belgium are not likely to reach their renewable energy goals set for next year (Eurostat,
2019). According to the directives set by the European Union, in Europe consensus exists
regarding the need to reduce greenhouse gas emissions. Failing to reduce greenhouse gas
emissions may result in accelerated climate change associated with significant economic
and social damages (Nordhaus, 2006). Therefore, it is expected that the share of energy
from renewable resources as a percentage of gross final energy consumption will further
increase in the coming years.
The worldwide electricity generation was 22 504 TWh in 20121 (Liu, 2015). According
to the base scenario “New Policies” of the International Energy Agency (Tanaka et al.,
2010), electricity generation will grow to around 36 600 TWh by 2035. Conti et al. (2016)
predict that total generation from renewable resources will increase by 2.9 percent per
year, as the renewable share of world electricity generation grows from 22 percent in
2012 to 29 percent in 2040. Total expenditure on energy was over 6 trillion US dollar in
2011, equal to around 10 percent of the world gross domestic product (Desbrosses, 2011).
Altogether, the increasing impact of renewable energy on the world economy indicates
the relevance of research in this area.
It is unknown what effect renewable energy sources will have on the future electricity
price. For firms in need of electricity it is important to use financial instruments like
forward rate agreements. In a forward rate agreement, investors can secure a price today
for the amount of energy delivered later on. The current energy price on the forward
energy market reflects expected prices, the question arises what the effect of increasing
renewable energy will be on the electricity price. Moreover, how can investors anticipate
this using forward contracts? Besides the question whether forward electricity prices
in-crease or dein-crease, it is important to consider other effects on the price like volatility.
According to Mulder and Scholtens (2016) price volatility will increase due to the
influ-ence of weather circumstances. Therefore, this paper studies the effect of the change in
renewable energy production on the forward electricity market and the following research
question is formulated:
Is the forward electricity price in the European market influenced by changes in the
share of renewable energy production?
In Europe, national electricity markets are intensively connected for more efficient
allocation of cross-border capacity. Besides physical grid connections, the financial
elec-tricity market is highly connected between the countries. Moreover, technologies to
im-prove storability and increased efficiency of the balancing financial market will lead to less
friction costs and thus lower prices. Renewable energy sources are able to reduce the
Eu-ropean Union dependence on foreign energy imports, also meeting sustainable objectives
to tackle climate change and to enhance economic opportunities (Cucchiella, D’Adamo,
& Gastaldi, 2018).
In this thesis a panel data analysis of eight European countries from 2010 to 2018
is performed. Data from Bloomberg Finance (Bloomberg L.P., 2019) and the European
Network of Transmission System Operators for Electricity is used to evaluate the effect
of the share of renewables in the energy mix with respect to the forward electricity price.
2
Literature Review
Renewable energy is defined as energy from renewable non-fossil sources, namely wind,
solar, aerothermal, geothermal, hydrothermal and ocean energy, hydropower, biomass,
landfill gas, sewage treatment plant gas and biogases (European Union, 2009). The
the merit order effect. The merit order is the order of electricity sources ranked by
their relative short-run marginal costs (Sensfuß, Ragwitz, & Genoese, 2008). Since wind
and solar energy generation have low short-run marginal costs, these sources are at the
beginning of the merit order. After these low-cost renewables, the merit order follows with
nuclear energy, lignite, hard coal, gas and fuel oil plants (Cludius, Hermann, Matthes, &
Graichen, 2014).
Sensfuß et al. (2008) state that renewable electricity generation has a considerable
impact on market prices. Alternatively, Mulder and Scholtens (2013) do not find an
increasing influence of renewable energy sources on the electricity price in the Netherlands.
However, they state that this may change when the magnitude of renewable energy
production continues to grow. Electricity prices may become less related to the marginal
costs of conventional power plants. Instead, they might increasingly be driven by weather
conditions on the one hand and scarcity in peak supply on the other hand.
Cludius et al. (2014) found that increased renewable energy production reduced
elec-tricity spot prices in Germany by six euro per mega watt hour in 2010, 10 AC/MWh in 2012 and with a near-term forecasting tool, they suggest that the price reduction should
have been 14-16AC/MWh in 2016. Mulder and Scholtens (2016) argue that the renewable energy producers connected to the German electricity market have a moderate impact on
the Dutch electricity market. When renewable electricity production is high in Germany,
the day-ahead electricity prices in the Dutch market are reduced. Sometimes even
nega-tive prices occur because of renewable energy costs not responding to wholesale prices of
conventional plants. The price pattern is flatter during the day as solar produces during
peak load hours. Moreover, there is higher price volatility from day-to-day and
week-to-week because of the influence of weather circumstances. Prices and volumes in balancing
markets do need to change dramatically as flexibility can be contracted in day-ahead and
intra-day markets.
Overall, it is assumed that a higher amount of renewable energy production lowers
the forward prices of electricity. Thus, a negative correlation between the forward prices
Mulder and Scholtens (2013) find that an increase in the deployment of renewable
energy sources leads to electricity markets being more sensitive to weather conditions.
Moreover, new technologies will come in and be more efficient so prices will be fluctuating
because of the varying efficiencies of the set of plant being used for generation at any
particular moment in time. Therefore, it is assumed that renewable energy production
will increase price volatility in the forward market.
3
Forward Electricity Market
In this thesis, the effect of renewable energy production on the forward electricity price will
be investigated. Participants in electricity markets attempt to minimize their exposure
to price volatility. Investors can reduce their exposure to price volatility by participating
in the forward markets. Ausubel and Cramton (2010) state that the forward markets
reduce transaction costs and improve liquidity and transparency. The forward markets
reduce risk, mitigate market power, and coordinate new investment.
A trade in the forward market can be either an agreement to take physical delivery of
the good or service at a given time in the future at a fixed price or alternatively a trade
in the forward market may represent a financial agreement to pay the difference between
the agreed forward price and the wholesale spot price at the designated time in the future
(Biggar & Hesamzadeh, 2014). Unlike in financial markets, the location of delivery plays
an important role in commodity trading, since transportation can be costly or dependent
on access to a grid. Therefore, commodity prices are usually quoted with reference to the
delivery point.
Large electricity market participants often seek to hedge their risk using risk-sharing
contracts. These risk-sharing contracts are often traded between market participants
or on organized securities exchanges, however there is not always a match between the
buyers and sellers of these hedge contracts. Especially when the market participants are
located at different geographical locations. In this case, there can be a role for a forward
the contractual forms. Forwards can be divided into individual power schedules and
standard forwards. Individual schedules are delivery schedules, during which power can
vary every hour or even every balancing period. Standard contracts are baseload or
peakload contracts which have delivery periods of either a day, week, month, quarter or
year. The baseload is set by the most efficient plants, with the lowest marginal costs
and will operate most of the time. During periods of high demand, the peak power
plants, which are relatively expensive will be used. Generally, peakload plants will only
be operating for a few hours (Bunn, 2004).
The Day-Ahead electricity market is a Day-Ahead hourly market for electricity that
allows market participants to schedule energy sales, purchases, and deliveries at binding
Day-Ahead prices for each hour. The Real-Time electricity market is a balancing market
that allows participants to coordinate the continuous buying, selling, and delivery of
electricity (Jones, 2017).
The most common form of hedge contracts base the financial payment on the
ob-served spot market price which is usually highly correlated with the profit of the market
participants. It provides important signals about future expected market prices,
allow-ing market participants to make investment decisions about future average price levels.
Potential problems with forward transactions is that their value depends on the
finan-cial capacity of the trading partner to complete the transaction. A promise to make a
financial payment in the future is of little value if, at the point in time when the
pay-ment must be made, the trading partner is insolvent. Therefore, in the case of direct or
over-the-counter trade, the parties will pay particular attention to the credit-worthiness
of their counterparties.
Erni (2012) observes characteristics of electricity price patterns such as seasonality,
high and clustered volatility, or extreme price observations. Given the increasing
pro-motion of electricity from renewable sources and according amendments in renewable
energy law, special attention is paid to the impact of expected wind electricity infeed and
its explanatory power in forecasting electricity spot prices. Moreover, markets must be
forecasts into markets. The prices for energy products need to reflect the expected
vari-ability in renewable generation as well as its forecast uncertainty. Furthermore, market
participants must be able to hedge against the possibility of more volatility in short-term
prices and to recover both their fixed and variable costs from their revenue stream offered
in the electricity market.
4
Methodology
The outcome of interest in this thesis is the relationship between renewable energy
pro-duction and the forward electricity price. At first, this relationship is tested for all
separate countries in the dataset. Establishing this relationship without panel analysis
is not sufficient to answer the research question. However, it helps to understand the
further analyses and to provide initial insight. In the following subsection, the panel
data analysis incorporates several control variables and alternative estimation methods
to answer the objective of this study. Control variables are needed to account for the
different demographic features per country. Adding control variables to the regression
removes the effect the control variables have on the relationship of interest between the
renewable electricity generation and the forward electricity price. The volatility
anal-ysis adds evidence to the effect renewable energy production share has on the forward
electricity price.
The forward year prices for electricity in this study are based on baseload contracts.
The prices for baseload rather than peakload contracts are used, because renewable
elec-tricity primarily determines the price for baseload contracts. Peakload prices are mainly
related to peak power plants with higher marginal costs in the merit-order of production
and operate during periods of high demand (de Miera, del Río González, & Vizcaíno,
2008). The share of renewable electricity production is calculated by dividing the net
renewable generation with respect to total net generation. The dataset transformations
4.1
Separate country analysis
In the first part of the statistical analysis, the simplified regression considers countries
separately and without any control variables to estimate the relationship between
re-newable energy production and the forward electricity price. Equation (1) shows the
regression for separate countries (i) and periods (t).
Fit= β0+ β1∗ Rit+ it (1)
In this model F is the baseload power one year forward price. The constant β0 is the
intercept of the regression. The variable of interest R indicates renewable energy as a
share of total production in each country. The error-term includes all determinants of
the dependent variable not given by the independent variables.
4.2
Fixed effects analysis
The availability of repeated observations on the same units allows to specify and estimate
more complicated and more realistic models than a single cross-section or a single time
series would do (Verbeek, 2008). Panel data make it possible to analyze changes on
an individual level. That is, panel data are not only suitable to model or explain why
individual units behave differently but also to model why a given unit behaves differently
at different time periods.
To test for a causal relationship between renewable energy production in a panel data
estimation, the fixed effect method proposed by Angrist and Pischke (2008) allows to
control for the different characteristics between countries and deals with the unobserved
effects in determining the forward prices. Fixed effects analysis relies on comparisons in
levels and assumes the trend behavior of the countries to be the same. The individual
averages for each country and variable are calculated over time and subtracted from
the observations. Control variables are needed to remove the effect the general regression
removes the effect the control variables have on the change in the β1 coefficient of interest.
to better inference. The regression is shown in equation (2).
Fit = β0+ β1∗ Rit+ β2 ∗ Dit+ β3∗ Git+ it (2)
The first control variable Ditmeasures the level of electricity demand. Daily day-ahead
natural gas prices in variable Git are a measure for the marginal costs of production.
4.3
Volatility analysis
In the volatility analysis the standard deviation of the price of the contracts is calculated
as a dependent variable and regressed upon the independent variables. This analysis
shows whether the production of renewable energy impacts the magnitude of the change
in the forward contract prices.
The volatility index is calculated using the methodology set out by European Union
(2016) and shown in equations (3) and (4). From the daily forward electricity prices (Pt),
the logarithmic difference of two consecutive trading days are computed (xi) in equation
(3). The mean (¯x) is calculated using the number of trading days (n) per month.
xi = log(Pt) − log(Pt−1) , x =¯ P
(xi)
n (3)
Using these figures, the standard deviation is calculated for each month. The value
obtained is multiplied by the square root of the number of trading days in that month
and a factor 100 to arrive at monthly percentage values. Weekend prices are eliminated
from volatility calculations, because lower trading volumes and as a consequence higher
daily price variations bring a bias into the volatility values. For this reason an average
monthly 21 trading day time period is taken for the volatility computations.
V olatility = s P (xi− ¯x)2 n ∗ 100 ∗ √ n (4)
The computed volatility of the forward electricity prices is used in regression (5). For
Vit = β0+ β1∗ Rit+ β2∗ Dit+ β3∗ Git+ it (5)
In this regression Vit is the level of volatility, Rit the share of renewable production,
Dit the demand with respect to capacity and Git the gas price.
5
Data
The dataset used in this study includes information on forward electricity prices and
renewable energy production. In addition, gas price and electricity demand are used to
identify the relationship between forward electricity prices and the share of renewable
energy production.
The dataset consists of daily price observations from 2010 until the end of 2018 for
all trading days. All data is adjusted to monthly intervals as this is the most frequent
time interval available for renewable energy production per country. The eight European
countries in the study are: The Netherlands, Germany, Switzerland, Czech Republic,
France, Italy, Belgium and Hungary.
Table 1: Aggregate descriptive statistics
This table summarizes the statistics of the dataset for all countries. The countries in this study are: The Netherlands, Germany, Switzerland, Czech Republic, France, Italy, Belgium and Hungary. The dataset covers monthly data from January 2010 through December 2018.
Obs. Mean Std. Dev. Min Max Forward year price (AC/MWh) 853 45.85 10.78 21.68 77.53 Renewable energy production (MWh) 864 3.087e+06 4.270e+06 1.9e+04 2.104e+07 Load (MW) 864 2.355e+04 2.116e+04 4127 7.820e+04 Gas price (AC/MWh) 635 27.63 12.99 11.98 81.46
The data on forward energy prices are derived from Bloomberg Finance (Bloomberg
L.P., 2019) The forward price agreements for a one year period are used to overcome the
problem of seasonality and daily volatility influencing the potential effect of renewable
energy generation. For most, but not all periods and for all countries the gas prices
measured in euro per mega watt hour are included as well and retrieved from the Thomson
Reuters Eikon data service (Reuters, 2019).
Transmis-Figure 1: Forward year electricitity prices
This figure shows the forward year electricity prices in euro per mega watt hour for the Netherlands (NL), Germany (DE) and France (FR) during the period from 2010 to 2018.
sion System Operators for Electricity (ENTSO-E, 2019). This association of 41
Trans-mission System Operators from 34 countries coordinates overarching grid topics to ensure
a constant power frequency in the transmission system. The ENTSO-E was established
in 2008 as the successor of the European Transmission System Operators (ETSO) and
publicate monthly historical data for aggregate statistics per country. Renewable energy
production is measured in MWh for the level of renewable energy capacity per country.
According to the merit-order of electricity, capacity of production determines the price.
However, information on capacity is not available on monthly intervals. Therefore, the
share of renewable electricity with respect to the total production of electricity is used as
indicator for capacity. The renewable energy production is characterized by its short run
marginal costs, so the level of production is a good indicator of capacity at the beginning
of the merit order.
The demand for electricity is measured via the electricity load measured at hourly
intervals. The maximum hourly load during the month is considered as a proxy for the
capacity. Dividing the actual load by the capacity and gives an indication of the relative
for analysis. In the Appendix in table 6 the descriptive statistics of each variable per
country are displayed.
The data for energy demand is used as a higher demand for electricity implies that
the demand curve shifts to the right along the merit order, enabling the marginal firm to
charge a price above its marginal costs until the marginal costs of the last unit not
dis-patched (Wilson, 2000). The shape of both the supply and demand curve are important.
The flatter these curves, the smaller the price effect of changes in supply and demand,
the higher the price elasticity. If demand is relatively high, less capacity is available
to respond to further demand increases. This makes supply more inelastic in the short
term. The net load is the amount that is on the grid at any time and is closely related to
electricity demand. The load on the grid has to be adjusted to meet demand as electricity
can not be stored easily. Therefore, an increase in the net load reflects a rightward shift
of the demand curve. Assuming the merit-order curve is constant, a rightward shift of
demand increases electricity prices.
Gas-fired power plants determine a significant part of the merit order in the market
An increase in the gas price means that the supply curve moves upwards. According to
Mulder and Scholtens (2013) the gas price appears to be a key factor behind electricity
prices, with an elasticity of about 0.6, this exceeds the impact of changes in demand in
their study.
Electricity is often traded on exchanges close to an hour before it is needed, in this
short term, the variable cost of power generation is essentially just the cost of natural
gas. Even in electricity systems with a substantial share of hydro and nuclear, the fossil
fuel plant often dictates the market prices. With increasing usage of gas resources for
electricity generation, in some markets the issue of whether gas drives electricity prices, or
vice versa, is not easily answered. With convergence between gas and electricity markets,
6
Results
The results of the seperate country regression (1), testing the relationship between
re-newable energy and the forward price, are shown in table 2 and appendix table 5. Once a
relationship has been established, the forward price effects due to renewable energy
pro-duction are tested as a panel. Finally, the results of the volatility analysis are presented
in table 4.
6.1
Separate country regression
In table 2 the results of the separate country regression are shown. For space
considera-tions only the results of the first three countries, The Netherlands, Germany and France,
are shown. The regression results for Switzerland, Czech Republic, Italy, Belgium and
Hungary are shown in the appendix in table 5.
Table 2: Separate country regression results
This table summarizes the results from regression (1): Fit= β0+ β1∗ Rit+ it. Standard errors are between parentheses. ***, ** and * denote the statistical significance at the 1, 5 and 10 percent levels respectively. Model (1) represents the Netherlands, (2) Germany and (3) France.
(1) (2) (3) β1 0.146 -0.695*** -0.598*** (0.240) (0.096) (0.010) β0 42.79*** 57.14*** 50.34*** (3.221) (2.439) (1.137) Observations 108 108 108 R2 0.004 0.333 0.254 F-statistic 0.37 52.84*** 36.08***
As can be seen from table 2 and table 4 in the appendix, all countries except the
Netherlands have a significant negative relationship at the one percent level between the
production of renewable energy and the forward electricity price. For the Netherlands
it is not possible to interpret the sign of the coefficient as it is not significant. For
Germany and France the coefficient can be interpreted as follows: a one percent increase
in renewable electricity production decreases the forward electricity price with 0.146 euro.
Given that the mean value from table 6 of the forward year price is 40.37 for Germany, the
of the regression. The constant can be interpreted as the intercept of the regression. For
Germany the expected average value of the Forward electricity price equalsAC57.14/MWh in case the renewable share has a value of zero.
Looking at the R2 of the independent regressions, with 0.333 for Germany and France
the overall explanatory power of the regression is not very high. This makes it necessary to
incorporate more variables in the regression equation that explain the forward electricity
price. The significant results from table 2 establish a relationship between renewable
energy production and the forward electricity price.
6.2
Fixed effects results
The Hausman test2 confirmed that the null hypothesis, which states that the difference
in coefficients is not systematic, can be rejected. Therefore, the fixed effects method is
adopted in regression (2) and the results are shown in table 3.
Table 3: Fixed effects regression results
This table summarizes the results from regression (2): Fit = β0 + β1 ∗ Rit + β2 ∗ Dit + β3 ∗ Git + it. Standard errors are between parentheses. ***, ** and * denote the
statisti-cal significance at the 1, 5 and 10 percent levels respectively. The time fixed effects dum-mies are included in the regression but excluded from the table for space considerations.
(1) (2) β1 -0.069*** -0.277*** (0.024) (0.047) β2 -0.040 -0.418*** (0.049) (0.111) β3 -0.142*** 0.639*** (0.024) (0.044) β0 58.17*** 63.53*** (4.069) (8.969) Observations 624 624 R2 0.705 0.164 F-statistic 74.30*** 115.7***
Time-fixed effects Yes No
Considering the country-fixed effects and time-fixed effects model (1), the coefficient
of -0.069 means that a one percent increase in the share of renewable energy production
decreases the forward electricity price with 0.069 euro per MWh. Additionally, the gas
price control variable is significant at the one percent level. An increase in the gas price
of one euro per MWh decreases the forward electricity price by 0.142 euro per MWh. The
constant can be interpreted as the intercept of the regression. Addition of the control
variables increases the R2 of the regression to 0.705. Thereby, the overall explanatory power of the model to estimate the effect renewable energy has on the forward electricity
price increases.
For model (2) the time-fixed effects are not included. The forward electricity price
decreases by 0.277 euro per MWh for each percentage point increase in the share of
renewable electricity production. The control variables in the regression for the forward
year contract can be interpreted as they are all significant at the one percent level. An
increase in demand with one percent decreases the forward price by 0.418 euro per MWh.
Furthermore, an increase of one euro per MWh in the gas price increases the forward
electricity price by 0.639 euro per MWh. The R2 of model (2) decreases in comparison
to model (1) which includes additional time-fixed effect dummies in the model.
6.3
Volatility analysis results
The results of the volatility analysis regression are shown in table 4. The renewable energy
production with a coefficient of 0.014 is not significant with respect to the price volatility
of the forward year contract. Therefore, it is not possible to infer a relationship between
the renewable energy production and the forward price volatility. The gas price with a
coefficient of -0.028 is significant at the ten percent level. An increase in the gas price of
one euro per MWh decreases the level of volatility with 0.028 percent. This effect is very
limited in magnitude, given the average gas price of 27.63 from table 1. For the model
without time-fixed effects there is a positive relationship between the share of renewable
electricity production and volatility of the forward energy price. The coefficient of 0.028 is
Table 4: Volatility analysis regression results
This table summarizes the results from regression (5): Vit = β0 + β1 ∗ Rit + β2 ∗ Dit +
β3 ∗ Git + it. Standard errors are between parentheses. ***, ** and * denote the
statisti-cal significance at the 1, 5 and 10 percent levels respectively. The time fixed effects dum-mies are included in the regression but excluded from the table for space considerations.
(1) (2) β1 0.014 0.028** (0.013) (0.011) β2 3.590 8.169*** (2.778) (2.692) β3 -0.028* -0.041*** (0.014) (0.011) β0 5.172 -3.512 (2.283) (2.176) Observations 608 608 R2 0.326 0.000 F-statistic 5.25*** 13.05***
Time-fixed effects Yes No
increases the level of volatility with 0.028 percent. The control variables for the level of
demand and the gas price are significant at the one percent level. However, given the low
R2 this model does not provide strong estimates for the relationship between renewable energy production and the level of volatility.
7
Discussion
The main limitation of this study is that the dataset does not contain observations for
all variables during all periods and for all countries. The forward prices retrieved via
Bloomberg Finance (Bloomberg L.P., 2019) are only available for eight European
coun-tries. Therefore, the panel analysis is limited to the variation between these councoun-tries.
Ideally, forward electricity price data from all European markets would be used to
im-prove the power of the tests. Moreover, the share of renewable energy with respect to
total generation is used to identify the effect on the forward electricity price. This is
an indication of the share of renewable electricity to capacity, which is not available at
monthly time intervals. Furthermore, the level of electricity demand is measured
Switzerland which limits the power of the panel analyses in equation (2) and equation
(2).
The results from table 4 show that volatility of the forward price is not significantly
influenced the share of renewable electricity production. The level of volatility due to
renewable energy production is probably limited since the forward year contracts are
not influenced by the seasonality factor of electricity prices. Forward year contracts for
shorter time periods might show more volatile price behaviour during the year.
8
Conclusion
This thesis studies the effect the impact of renewable energy production has on the
forward electricity price. The effect of the changing share of renewables in the energy
mix is estimated on prices and the volatility level in the forward market from 2010 to
2018. The dataset covers eight European countries and provides information on the
forward electricity price, the share of renewable energy production, the level of electricity
demand and the natural gas price. The findings from separate country analyses indicate
a negative relationship between the share of renewables and the forward electricity price.
This negative relationship is confirmed by panel analyses using a fixed effects estimation.
The results of this study suggest a negative relationship between renewable energy
production and the forward electricity price. The increasing share of renewable energy
production lowers the price investors pay for expected electricity in a year. The forward
electricity price in the European market is thus influenced by changes in the share of
renewable energy production. However, it can be concluded that the effect is limited in
size. In addition, the level of volatility of the forward price does not seem to be related
to the share of renewable energy production. Therefore, for investors it does not become
more difficult to hedge against the expected electricity price due to renewable energy
production.
The limitations of this study could largely be resolved by monitoring information
frequent level. Moreover, increasing the number of countries in the study and researching
other time periods might provide interesting insight. In addition, other countries outside
Europe where renewable electricity is emerging could be studied for more robust results
regarding the relationship with the forward electricity price. Furthermore, forward
con-tracts with different maturities can be assessed to check if the results are present for these
contracts as well and whether this impacts the level of volatility. Together with the aim
References
Angrist, J. D. & Pischke, J.-S. (2008). Mostly harmless econometrics: An empiricist’s
companion. Princeton university press.
Ausubel, L. M. & Cramton, P. (2010). Using forward markets to improve electricity market design. Utilities Policy, 18 (4), 195–200.
Biggar, D. R. & Hesamzadeh, M. R. (2014). The economics of electricity markets. John Wiley & Sons.
Bloomberg L.P. (2019). Baseload electricity forward prices for delivery on high voltage grid. Retrieved from Bloombergterminal
Bunn, D. W. (2004). Modelling prices in competitive electricity markets.
Cludius, J., Hermann, H., Matthes, F. C., & Graichen, V. (2014). The merit order effect of wind and photovoltaic electricity generation in germany 2008–2016: Estimation and distributional implications. Energy Economics, 44, 302–313.
Conti, J., Holtberg, P., Diefenderfer, J., LaRose, A., Turnure, J. T., & Westfall, L. (2016).
International energy outlook 2016 with projections to 2040. USDOE Energy
Infor-mation Administration (EIA), Washington, DC (United States . . .
Cucchiella, F., D’Adamo, I., & Gastaldi, M. (2018). Future trajectories of renewable energy consumption in the european union. Resources, 7 (1), 10.
de Miera, G. S., del Río González, P., & Vizcaíno, I. (2008). Analysing the impact of renewable electricity support schemes on power prices: The case of wind electricity in spain. Energy Policy, 36 (9), 3345–3359.
Desbrosses, N. (2011). World energy expenditures. Leonardo Energy, 28. ENTSO-E. (2019). Retrieved from https://www.entsoe.eu/data/power-stats/
Erni, D. (2012). Day-ahead electricity spot prices: Fundamental modelling and the role of
expected wind electricity infeed at the european energy exchange. na.
European Union. (2009). Directive 2009/28/ec of the european parliament and of the council of 23 april 2009 on the promotion of the use of energy from renewable sources and amending and subsequently repealing directives 2001/77/ec and 2003/30/ec.
Official Journal of the European Union, 5, 2009.
European Union. (2016). Methodological description and interpretation of the volatility index for electricity markets. Quarterly report on European electricity markets. Re-trieved from https://ec.europa.eu/energy/sites/ener/files/documents/volatility_ methodology.pdf
Eurostat. (2019). Share of renewables in energy consumption in the eu reached 17,5%
in 2017. Office for Official Publications. Retrieved from https : / / ec . europa . eu /
eurostat/documents/2995521/9571695/8- 12022019- AP- EN.pdf/b7d237c1- ccea-4adc-a0ba-45e13602b428
Jones, L. E. (2017). Renewable energy integration: Practical management of variability,
uncertainty, and flexibility in power grids. Academic Press.
Liu, Z. (2015). Global energy interconnection. Academic Press.
Meurink, A., Muller, G., & Segers, R. (2017). Hernieuwbare energie in nederland 2017.
Den Haag/Heerlen/Bonaire: Centraal Bureau voor de Statistiek.
Mulder, M. & Scholtens, B. (2013). The impact of renewable energy on electricity prices in the netherlands. Renewable energy, 57, 94–100.
Nordhaus, W. D. (2006). Geography and macroeconomics: New data and new findings.
Proceedings of the National Academy of Sciences, 103 (10), 3510–3517.
Reuters, T. (2019). Thomson reuters eikon. Thomson Reuters.
Sensfuß, F., Ragwitz, M., & Genoese, M. (2008). The merit-order effect: A detailed anal-ysis of the price effect of renewable electricity generation on spot market prices in germany. Energy policy, 36 (8), 3086–3094.
Tanaka, N. et al. (2010). World energy outlook 2010. International Energy Agency. Paris:
IEA.
Verbeek, M. (2008). A guide to modern econometrics. John Wiley & Sons.
9
Appendix
Table 5: Separate country regression results
This table summarizes the results from regression (1): Fit = β0 + β1 ∗ Rit + it. Standard errors are between parentheses. ***, ** and * denote the
statis-tical significance at the 1, 5 and 10 percent levels respectively. Model (4) rep-resents Switzerland, (5) Czech Republic, (6) Italy, (7) Belgium and (8) Hungary.
(4) (5) (6) (7) (8) β1 -0.104*** -8.28*** -0.685*** -0.507*** -0.864*** (0.036) (1.115) (0.061) (0.151) (0.237) β0 48.58*** 48.47*** 73.63*** 52.76*** 54.05*** (1.209) (1.376) (1.567) (2.347) 2.133) Observations 108 108 108 97 108 R2 0.073 0.342 0.540 0.107 0.112 F-statistic 8.39*** 55.19*** 124.5*** 11.34*** 13.30***
For Switzerland, the Czech Republic, Italy, Belgium and Hungary there is a significant
negative relationship at the one percent level between the forward electricity price and
the share of renewable electricity production. Particularly, for the Czech Republic the
relationship is pronounced. The coefficient of -8,28 indicates that for every percentage
point increase in the share of renewables, the forward electricity price decreases by 8.28
Table 6: Descriptive statistics per country
Obs. Mean St.Dev. Min Max
Netherlands
Forward year (AC/MWh) 108 44.68 8.556 25.75 60.72 Renewable (MWh) 108 1.118e+07 3.039e+05 1.46e+05 1.865e+07 Demand (MW) 108 1.285e+04 755.5 1.155e+04 1.438e+04 Gas (AC/MWh) 108 20.81 4.465 11.98 32.11
Germany
Forward year (AC/MWh) 108 40.37 10.02 21.68 59.08 Renewable (MWh) 108 1.169e+07 4.216e+06 5.220e+06 2.104e+06 Demand (MW) 108 5.625e+04 3513 4.926e+04 6.417 Gas (AC/MWh) 108 20.90 4.476 12.21 31.65
Switzerland
Forward year (AC/MWh) 108 46.55 10.59 25.11 65.98 Renewable (MWh) 108 1.036e+06 1.493e+06 1.04e+05 4.952e+06 Demand (MW) 108 6179 1080 4227 8260 Gas (AC/MWh) n.a. n.a. n.a. n.a. n.a.
Czech
Forward year (AC/MWh) 108 39.83 9.358 22.15 56.48 Renewable (MWh) 108 4.538e+05 2.727e+05 1.9e+04 9.72e+05 Demand (MW) 108 7309 734.2 6071 9177 Gas (AC/MWh) 44 18.68 4.165 12.34 28.11
France
Forward year (AC/MWh) 108 44.83 8.042 26.08 60.43 Renewable (MWh) 108 4.128e+06 3.074e+06 9.57e+05 1.173e+07 Demand (MW) 108 5.512e+04 1.0230e+04 4.158e+04 7.820 Gas (AC/MWh) 108 21.01 4.541 12.27 32.65
Italy
Forward year (AC/MWh) 108 58.53 12.04 37.44 77.53 Renewable (MWh) 108 5.129e+06 3.017e+06 1.072e+06 1.168e+07 Demand (MW) 108 3.629e+04 1991 3.182e+04 4.299e+04 Gas (AC/MWh) 62 21.44 4.180 13.66 30.04
Belgium
Forward year (AC/MWh) 97 45.27 7.816 28.09 64
Renewable (MWh) 108 9.389e+05 2.593e+05 5.13e+05 1.463e+06 Demand (MW) 108 9792 763.3 8439 1.172 Gas (AC/MWh) 97 51.69 11.76 27.93 81.46
Hungary
