Energy transitions in Built Environment of Netherlands: A
System Dynamics approach to diffusion of Solar boilers and
Insulation
Tariq Abdul Muhaimin Mavroeidis Nikolaos
Faculty of Technology, Policy and Management
Delft University of Technology
Jaffalaan 5, 2628 BX, Delft
Telephone: + 31 15 27 89801
tarig.abdul.m@ gmail.com ; mavroeidis.n@ gmail.com
Abstract
In context of rising demand for energy amidst limited resources, energy
efficiency is one of the major concerns of a modernized world. Gas consumption in
the built environment constitutes for more than 30% of the overall energy
consumption in the world. Hence, energy transitions and their preferences in the built
environment demand urgent attention by policy makers in order to implement cost-
effective and long-term sustainable policies. However, the highly dynamic nature of
decision making adopted by households involves complex interaction between various
factors. Therefore, in this study, a modified Bass diffusion structure was used in a
System Dynamics model to examine energy transitions and subsequent reductions in
overall gas consumption in the built environment of Netherlands. Specifically, the gas
consumption by different types of owner-occupied houses was studied by considering
the probability of adoption of solar boilers and/or insulation by these households.
Also, the effect of different policy measures such as subsidies, demolition, innovation
driving, and increased advertisement/awareness, under different scenarios, was
evaluated. These experiments confirmed that there is great potential for energy saving
in the building sector. Moreover, policies that focus on changing people's behavior
were found to be more effective in the energy transitions process.
Keywords: System Dynamics, energy transition, gas consumption, solar boilers,
insulation, Built Environment, Netherlands
1. Introduction
In the last two decades, issues such as global climate change, security of fossil
fuel supply, and potential risk of future fossil fuel scarcity have led to discussions and
policies aiming for transitions to energy sustainability at both national and
international levels. In March 2007, the Brussels European Council acknowledged the
importance of energy conservation and called on European Union (EU) member states
to pursue actions to develop a sustainable integrated European climate and energy
policy (Council of EU, 2007). In their statement, EU leaders made a firm independent
commitment to achieve at least 20% reduction in greenhouse gas emissions by 2020
compared to 1990. They also endorsed the EU aim of having a binding target of a
20% share of renewable energies in overall EU energy consumption by 2020.
In northern European countries such as The Netherlands, built environment
contributes to about 30% of the total energy consumption (Ministry of the Interior and
Kingdom Relations, 2011; ING, 2013). Moreover, a major part of the total Energy
consumption by households takes place in the form of natural gas consumption for
the purposes of water and space heating. Acknowledging this great potential for
energy saving, the Dutch Ministry of Interior and Kingdom Relations published the
Plan of Action: Energy Saving in Built Environment in 2011. In this plan, the Ministry
concluded that “by changing heating behavior, or installing insulation or improving
insulation, users or owners can ensure less energy is consumed” (Ministry of the
Interior and Kingdom Relations, 2011).
Till date, several researchers have tried to analyse the impacts of policies related
to energy transitions in the built environment. While some researchers focussed on the
prevailing inertia in the built environment sector due to old-age dwellings (Y ucel,
2013), others have studied the impact of energy-efficient innovations like building
designs on the overall energy consumption (Grosser et al., 2006). A few others have
also evaluated the impact of energy performance indicators like energy labels adopted
under the Energy Performance of Building Directive (EPBD) (BPIE, 2010; Beerepoot,
2007). But, direct evaluation of policies that aim at impacting household behavior
through the adoption of energy efficient measures like solar boilers and/or insulation
of buildings is missing.
In this paper, the effectiveness of four policies that aim to promote the adoption of
solar boilers and/or insulation measures by Dutch households is studied. The policies
selected for this study are: demolition of old-age dwellings and construction of new
insulated dwellings, provision of income-based subsidies for solar boilers and
insulation, investment in solar boiler and insulation related R&D, and
advertisement/awareness creation. Since the system involves complex interrelations
between different factors like dwelling age, new constructions, demolitions,
renovations and occupant behavior, a System Dynamics model was used for this study
in order to capture important feedback effects.
The paper is structured as follows: the following section introduces the scope and
structure of the simulation model. Then, Section 3 elaborates the simulation
experiments that were performed on the SD model. This includes an impact study of
the four policies. Section 4 is dedicated to general conclusions and indications for
possible future research.
2. Model description
The objective of the model is to simulate the aggregate natural gas consumption in
owner-occupied residential buildings of The Netherlands. Till date, about 56% of the
Dutch housing stock is owned by occupant-owners, and it is predicted that this
percentage will increase to 70% in the near future (Bogerd et al., 2009). Hence, a
commitment towards energy-saving among owner-occupied households is highly
important to achieve the targeted reduction in CO2 emissions. The model focuses on
the energy consumed for domestic water and space heating in dwellings, since this
constitutes a major portion (i.e., 70-80%) of the total residential energy consumption
(Yucel, 2013). The natural gas consumption by households depends mainly on the
dwelling type, dwelling age, and occupant’s response to various socio-economic
factors. Accordingly, the factors that are used to formalize the decision making
process by households (in regards to adoption of energy efficient measures) are shown
in Figure 1.
+ Dwelling type
Independent
+Row/Terraced
+ Apartment
Dwelling age
+Early (0-20)
+ Middle (20-40)
+Later (40+)
+ Climate
+ Income
+ Energy expenses
+ Efficiency of
Technology
+ Implementation &
maintenance costs
Imitation
+ Advertisement
Socio-
economic
factors
Figure 1: Factors involved in energy transition decision making by households
In brief, socio-economic factors (e.g. income, energy expenses, etc.) as well as
climatic conditions influence the natural gas demand of a household in the model.
This demand coupled with the quality of the dwelling (determined by its age and type)
determines the final natural gas consumption by that dwelling. Hence, the System
Dynamics model that simulates the total natural gas consumption of the Dutch
housing sector covers the following aspects in order to simulate aggregate
consumption at the national level.
2.1. Dwelling stock
As shown in Figure 1, the quantity aspect of the housing stock (the number of
dwellings), and the quality aspect of the housing stock (energy efficiency of the
dwelling) were modelled by considering the subdivision of this stock into different
groups. Because of the vast diversity in the type of dwellings, homogenising the
Dutch housing stock may result in the overlooking of important dynamics. On the
other hand, trying to capture diversity in its full scale may result in a model that is of
little use due to its level of detail. Considering this trade-off, the Dutch housing stock
can be grouped under three main categories: Independent dwellings, row/terraced
dwellings, and apartments (see Figure 2). The total Dutch dwelling stock and the
Independent dwelling Row/Terraced dwelling Apartment/Gallery flat
Figure 2: Representative dwelling types
distribution of different types of dwellings are shown in Figures 3 and 4.
According to Yucel (2013), Gallery flats are predominantly social housing units,
which accommodate low-income families and detached dwellings are commonly
occupied by high-income households. This fact was used in the modelling process in
order to consider the differential decision making aspect of different income groups.
Distribution of the Dwelling stock
SB Independent houses
™@ Row/terraced houses
Appartments
Figure 3: Distribution of dwelling stock (2000) (Yucel, 2013)
Total Dutch Dwelling Stock (1995-2011)
~
nou
Million Dwellings
o
in
1995
1996
1997
oa eunmauenae
SESEEEREEREE
2010
2011
Figure 4: Total Dutch Dwelling Stock (Yucel, 2013)
While describing the housing stock, another very important aspect, with respect to
energy efficiency, is the Dwelling age. Capturing the age distribution of the dwelling
stock is important because important characteristics such as construction materials
used, technology employed etc., during construction of dwellings change based on the
time period in which it is built. Also, the age distribution of dwelling stock is not
uniform. This is primarily because of changes in demand for housing over the years.
The distribution of dwelling age in The Netherlands as of year 2000 is shown in
Figure 5.
Age distribution of dwellings
100
80
e 60
* 40
20
0
Independent _Row/terraced _ Appartments
40+ 36 42 30
™ 20-40 aa 33 43
0-20 31 25 27
Figure 5: Age distribution of Dutch dwelling stock (2000) (Yucel, 2013)
To capture these incumbent fluctuations in the number of dwellings with different
service times, the age distribution of dwellings is classified into three stages (or age
categories). The first two age groups are of a 20-year duration, while the later stage
category consists of buildings that are more than 40 years old. The decision of
dividing the stock into these durations was taken due to data availability. But before
doing this, the total Dutch dwelling stock is divided into four parts — Households with
no insulation or solar boilers, Households with only insulation and no solar boilers,
Households with no insulation and only solar boilers, and Households with both
insulation and solar boilers. Each of these parts are modelled as a separate stock. And
within each stock, the three age categories were introduced at the micro-dynamics
level (Fallah-Fini et al., 2013) using the subscripting function of Vensim.
Figure 6 shows the dwelling stock structure that was made for No-insulation and No-
solar households. Within this structure, an aging chain is modelled. The ‘years in each
age group’ variable was initialized to 20 and used only for the early and medium
stages. The age duration of the later stage category is determined by the demolition
rate for that particular type of household. Although theoretically the useful service
time of a dwelling ranges between 75 and 100 years, historical figures in The
Netherlands show that the fraction of dwellings that are being demolished annually
ranges between 0.15 and 0.35% (Yucel, 2013). Hence, when a rough estimation is
done, these figures suggest an average service time greater than 200 years. Based on
this data, the demolition rates of independent houses, Row houses, and apartments
were initialized to be 0.005, 0.008 and 0.01 respectively, indicating an average service
time of 200 years, 125 years and 100 years.
Figure 7 shows the conceptual model representing the movement of households from
one stock to another. The four blocks present at the four corners of this figure
represent the dwelling stock that was divided into households without Insulation or
Solar boilers (NI-NS), Households with Insulation but no Solar boilers (WI-NS),
Households without Insulation but with Solar boilers (NI-WS), and Households with
both Insulation and Solar boilers (WI-WS). Each block also shows the subdivision of
these households into the early stage, middle stage and later stage. Within each of
these stages, again the households were classified into four income groups — First
Quartile, Second Quartile, Third Quartile, and Fourth Quartile.
2.2. Income Groups
Based on the World Development Indicators’ database of The World Bank, the gross
national income (GNI) at purchasing power parity (PPP) per capita of Netherlands in
2010 was classified into the following five quintiles: €11994 for the first 20%, €20864
for the second 20%, €27209 for the third 20%, €36709 for the fourth 20%, and
€61045 for the fifth 20% (calculated using the average exchange rate of 2010: 1 $ =
0.755 €) (World Bank, 2014). However, in order to reduce complexity and
considering the fact that the results will probably overestimate income in poorer
quintiles and underestimate income in richer quintiles because the shares were
calculated by ranking households, and not persons, the representative incomes for the
four income quartiles in this model were set as: First quartile — €10000, Second
quartile — €20000, Third quartile — €40000, and Fourth quartile — €70000.
These income groups, along with the dwelling type, dwelling age and other socio-
economic factors were then used to model household behavior and decision making
process. T his has been explained in detail in the next subsection.
NENS dwellings of
each type and income
group
Demolition rate
Demolition of, [onstruction of
H-NENS HH
Years in each
age group Construction rate
‘onstruction
Households without|
Insulation without of NENS
Aging of Solar boilers
H-NENS|
Figure 6: Dwelling stock used for modelling No insulation-No Solar households
4 Households with no insulation oF solar
Households with only insulation
later stage }
i Early stage 1 | Early stage H
l } r «b> — [Middle stage] d
= - + i
Later stage J a |] |_Lriaterstage | |!
= Poly
Pal
incomeof | @y “=
i <7
pwelling 295 household |
Paley:
Subsidesincentives
a ange mam Mopten
— | Fo solar and/or
v easton
[Efficiency of
ES [psulation/soig ad
easibility dud
Innovation to dwelling
devin 8&0 type
Tarly stage «sf early stage
+ i =
[nae snes} 4 [ima saps]
"i i —t
f “Later stage
Households with insulation and solar
Households with only solar boilers
Figure 7: Conceptual Model
2.3. Household behavior and Decision making
The central structure of the conceptual model in Figure 7 shows all the factors that
were used to model the decision making process. As pointed out by other researchers
(Ben Maalla & Kunsch, 2008; Y ucel, 2013), the behavior of households with respect
to their reaction to energy expenditures is extremely important This factor was
modelled as the willingness of households to adopt insulation and/or solar boilers
based on the amount that they are willing to spend on energy related expenses; and
this differs for each income group. Apart from this, the technical factors related to the
feasibility of households to adopt either of the two technologies are also important
(Van Raaij & Verhallen, 1986). This feasibility, which is mainly based on the
household type and technical efficiency of each technology, was modelled as
Attractiveness of Insulation/Solar. Also, social influences play a very important role
6
in determining the adoption rate of a particular technology. Hence, the effect of
advertisements and influences through physical contact like mouth-to-mouth
discussions were modelled as Adoption from advertising and Adoption from word of
mouth. In the SD model, a modified Bass diffusion structure as shown by Sterman
(2000) was used to incorporate this adoption process. The equations used for the
adoption rate are given below:
AR =W - AoT -(Ad +WoM )
where
AR:Adoption Rate
W: Willingness to adopt
AoT: Attractiveness of Technology
Ad: Adoption from advertising
WoM: Adoption from Word of Mouth
Ad =Ef-.F-P
where
Ad:Adoption from advertising
F:Feasibility of technology per type and age of dwelling
P:Potential adopters
WoM =c-af-F pA
D
where
WoM: Adoption from Word of Mouth
c: contact rate
af: adoption fraction
F: Feasibility of technology per type and age of dwelling
P: Potential adopters
A: Adopters
D: Total Dwellings
The two primary feedback structures that drive adoption of a technology are shown in
the Causal loop diagram of Figure 8. The first loop depicts adoption by households
based on the percentage of income that they are currently spending on energy
expenses. The current spending is compared with the percentage of income that they
are willing to spend. This is a negative feedback loop because an increase in the
number of insulated/solar-adopted houses will decrease the overall aggregate gas
consumption and in tum decrease the average willingness to adopt. The second
important loop is the effect of implementation cost, maintenance cost and efficiency
on the adoption. These 3 factors together determine the attractiveness of technology.
With more houses adopting insulation and/or solar, the learning curve effect causes a
positive feedback loop to function here (Kersten et al., 2011; Madlyusz & Pém, 2013).
Thus, the overall adoption rate of a particular technology is a function of all these
7
factors. However, the values of several exogenous variables were determined based
on an approximation of available information/data. These have been indicated in the
following subsection.
natural gas price
ty income
ames Heating expenses
haar ed per household - ;
Heating expenses to
Fr income ratio
Household
beating demand
Non-adoption due to
satisfaction with heating
Es
expenses Perceived heating
Reference heating _®*Penses fo Income
Avatee energy expenses to income
ratio
noweholdp a rape
Advertisement
effectiveness
Feasibility
Houses without
a Willingness to adopt
insliosoler stmionperd wend | Adoptionrate —_insulation/solar
Houses with + ., imitation
insulation/solar
| +
A=! of technol
with increased expenses
insulaben/ er
Adoption by
implementation ost
reduction (A trantiveness of
= Implementation technology
Leaming rate cost. =| —
for cost. __>
Maintenence cost ~
a \ reduction
laintenance
cost
fliciency
singroveni + A
Leaming rate for
+ Efficiency
efficiency
Figure 8: Causal Loop Diagram for decision-making by households
2.4. Primary assumptions in the model
A few important constants that were assumed are: advertisement effectiveness factors,
implementation costs, natural gas price, learning rates of cost and efficiency, contact
rates, reference heating expenses to income ratio, feasibility of adoption based on
dwelling type, convincing degree, construction rates and demolition rates of different
types of houses. However, to prevent any bias, the model was subjected to sensitivity
testing for a wide range of values in these variables. The Key Performance Indicators
(KPIs) were not found to be behaviorally sensitive to +/- 10% changes in these
constants. Hence, the assumed values could be effectively used for studying the long-
term behavior of this system, if not for exact estimation of outcomes. But, as
mentioned before, the purpose of this model is only to observe the possible behavior
of the system as a whole under different policy measures.
2.5. Model boundaries
Before moving on to the simulation experiments, it is important to point out that the
model used for this study has certain well defined boundaries. Most importantly, it
considers only owner-occupied residential buildings because of their ability to take
decisions related to the adoption of either of these two technologies. The Bulls eye
diagram of Figure 9 shows the variables/parameters that are included and the ones
that are excluded for this study. Some technical aspects related to houses are kept
outside the model boundaries because of the non-relevance of such micro aspects
while studying the aggregate/overall behavior.
Social housing and zt
rental houses a
GHG emmissions -
Learning rates for Fr = Demolition rates fo?
rat ss each type of
efficiency S
Attractiveness Attractiveness household
z of insulation of Solar
Construction cost of / a A
houses f / ~~ Households without 5
if rasucseewe ) Lo Insulation and Solar \ \
loption rateby
/ [Willingness / Households with Households with’ \
imitation \
Energy labels | toadopt / Willingness \ Construction rates
ly Sol
J cca jf "sition Soar onlySolar \\ “sgadopt \ for each type of
| | \ \ household
| | | Households with Efficiency of yor |
i | / only insulation technologies \ | |
Industrial and |
tor \ \ eee Maintenance costs | |
attontvenss; | Seat OF hl of both technologies / | Averagegas |
\ technologies | |
\ / | consumption of ‘|
\ \ / J “cachtecet |
Building design | \ ‘THOUROUGHLY MODELLED ENDOGENOUS household
\ VARIABLES /
\\ Natural gas price Thos Z
Surface area and R-\ SUPERFICIALLY MODELLED ENDOGENOUS VARIABLES
ues of h
values of houses el ZZ swcomeat
# households per
quartile
Technical feasibility of
Types of solar adoption
EXOGENOUS VARIABLES
Types of insualtion
DELIBERATELY OMITTED VARIABLES
Figure 9: Bulls eye diagram of model boundaries
3. Simulation experiments
As mentioned in Section 1, the primary objective of this study is to gain a generic
insight into the adoption trend of two energy efficient technologies — Insulation and
Solar boilers — under four different policy measures. Accordingly, only relevant
results of different experiments performed on the model are produced here. The
variables that served as Key Performance Indicators (KPIs) in this model are: 1. Total
gas consumption (in m3), 2. Total households with/without insulation and/or solar
boilers, 3. Willingness to adopt insulation/solar boilers, and 4. Attractiveness of
insulation/solar boilers.
Prior to entering the experimentation phase, the model was also tested for validity
of both its structure and expected behavior in order to see if it serves the purpose at
hand. For this, the established verification and validation procedures proposed in the
field (Sterman, 2000) were used. It is concluded that the model serves fit for the
purpose at hand.
In this section, the baseline case is first established in order to demonstrate the
aggregate gas consumption trend without any policy interventions. Then, the impact
of four policies — demolition of old-age dwellings, provision of income-based
subsidies for solar boilers and insulation, investment in solar boiler and insulation
related R&D, and advertisement/awareness creation — is studied in comparison with
this baseline behavior.
3.1. Baseline case
The baseline/reference case is a completely hypothetical case that can help in drawing
a baseline for the gas consumption trend in the owner-occupied residential sector of
Netherlands. In order to arrive at this, it is assumed that households continue to adopt
either/both Insulation and/or Solar boilers at the rate at which they do at present
depending on the values presumed for different constants and parameters in the model.
Also, the dwellings continue to exist as they were built i.e., the rate of building new
Insulated houses or houses with Solar boilers remains the same throughout the
timeline of 2013 — 2050.
As shown in Figures 10 and 11, in the reference case the total owner occupied
dwellings increase while the overall gas consumption decreases. This is caused by the
fact that households that have implemented at least one of the two technologies will
eventually implement the other one too; and most new constructions are already
assumed to be insulated. Also, as shown in Figure 12, the first quartile households are
willing to adopt energy efficient technologies more than the other quartiles because
their heating expenses devour a large share of their total income. However, they do
not proceed to implementation because the attractiveness of a technology (as shown
in Figures 13 and 14) considers the cost aspects of implementation; and
implementation cost is a big obstacle for the first quartile households. But, the
attractiveness of solar boilers is higher than that of insulation and more equally spread
because of relatively lower implementation costs.
The fluctuations observed in the willingness to adopt a technology (as shown in
Figure 12) are caused due to the influence of climatic conditions on the perceived
need to adopt insulation/solar boilers by households. In the model, these fluctuations
in climatic conditions are introduced through the use of Heating Degree Days (HDD).
The willingness to adopt a second technology, if they already have one, is found to be
less than half of the willingness to implement it for the first time. However, through
imitation and advertisement there is a constant increase in this rate.
A sensitivity analysis on parameters like “contact rate”, “feasibility of technology”,
“learning rates” showed mainly numeric sensitivity. Scenario-testing on HDD and
natural gas price showed that the willingness to adopt, and therefore the energy
transitions, are comparatively lesser in the “low” value scenarios.
10
Therefore, without implementation of any policy the total gas consumption is found to
be diminishing. However, since a higher rate of transition is desired, four different
policies are explored. The following subsections discuss the impact of these policies.
Total dwellings Total consumption
41M 6B
4.05M. 5.75B
4M fesse pee @ 55B
3,95 M 5.25B
9M 5B
‘ 2013 2017 2021 2025 2029 2033 2037 2041 2045 204!
‘Time (Year)
2013 2017 2021 2025 2029 2033 2037 2041 2045 204:
Time (Year)
Total dwellings : BaseCase
Figure 10: Total Dutch dwelling stock (reference case)
Total consumption : BaseCase
Figure 11: Total gas consumption (reference case)
willingness to adopt insulation by NI-NS
0
2013 2017 2021 2025 2029 2033 2037 2041 2045 204!
Time (Year)
BaseCase
Bi
BaseCase
Bi
Figure 12: Will to adoptinsul byh holds without Insulation or Solar (based on income quartiles)
attractiveness of Insulation
0.6
0.45
2 os Nee e eee ees aeeee eee ee eee
0.15
0
2013 2017 2021 2025 2029 2033 2037 2041 2045 2049
Time (Year)
attractiveness of BaseCase
attractiveness of BaseCase
attractiveness ase
BaseC
attractiveness of Insulation| fourthquartile) : BaseCase
t oLiasal
(based on income quartiles)
Figure 13:
byh
11
attractiveness of solar boilers
dam
=
0
2013 2017 2021 «2025 +2029 ©2033 2037 2041 +2045 204!
‘Time (Year)
attractiveness of solar nlerffouthquntie] :BasCase
Figure 14: Attractiveness of Solar boilers by households (based on income quartiles)
3.2. Demolition policy
The reference case experiment is extended by introducing a policy that aims at
promoting the demolition of later stage dwellings that do not have insulation or solar
boilers. Also, the construction of dwellings with insulation is promoted. This policy
could be implemented by setting a minimum level of insulation for new constructions.
In the experiment carried out using the model, this policy sets a target of doubling the
demolition rate of dwellings that have no insulation or solar boilers between the years
2015 and 2035. The policy results in an increase of the demolition and construction
rates as shown in Figures 15 and 16. Only the demolition and construction trends of
apartments are shown in these two figures. After the implementation of this policy,
the aggregate gas consumption is found to decrease immensely (see Figure 17). This
is due to the change in composition of the dwelling stock. If this policy is
implemented, the total number of dwellings with no insulation or solar boilers will
decrease by a large extent and the number of dwellings with a certain degree of
insulation increase. When this happens, households become more energy efficient as
the preferred ambient temperature is maintained without the use of natural gas for
space heating.
demolition of H-NI-NS
10,000 —
. Tao
| ee tt!
7,500
u
i 5,000 |—
4 earn,
2,500 peEeeEEcoeaa Tr
Ce eee enn EEER en eeeee
Co
0
2013 2017 2021 2025 2029 2033 2037 2041 2045 2049
Time (Year)
demolition of HENLNS 1 dmolton 2
acl of HNL {eiitesrprntonsort st Senos? 255 sndcnston 2205
damoliion of INL 3 denaliion2
demollon of MENS [aerpprtmentsfutite Poly rmain 2 203 andcondnation 22038
Figure 15: Policy - Doubling demolition rate ofnon-insulated and non-solar houses over 20 years
12
construction of H-WI
4,000
3,000
2,000
housholdY ear
1,000
0
2013 2017 2021 2025 2029 2033 2037 2041 2045 2049
ime (Year)
"construction of HW" Policy demolition 2
eons of HWP ey pprinents sonar); Ply donoition 22035 ndconsation2 2035
“construction of H-W Policy demolition 2. 2035,
“cunsration of BAW Toalapportmentsfouthgurl] Poy dmolion 2 2035 nd consmion2 2035
Figure 16: Policy - Doubling construction rate ofinsulated houses over 20 years
Total consumption
6B
re
SSS
5.75B TT Seo
2 55B
5.25B
5B
2013. 2017 2021 2025 2029 2033 2037 2041 2045 2049
otal consumption: Policy!_demolition 2 2035 and construction 2 2035
Total ton: Base ase
Figure 17: Total gas consumption after implementation of demolition policy
Also, as shown in Figure 17, this policy helps in setting the trend for a long-term
impact on the aggregate gas consumption. Although the implementation of this policy
is politically debatable, setting certain minimum allowable insulation standards will
certainly facilitate the energy transitions process in the Built environment.
3.3. Subsidies policy
Another policy that is usually employed by government agencies while promoting
certain technology or policy, is the provision of subsidies. In recent years, a budget of
€121 million was made available under the Ministry of Housing, Spatial Planning &
the Environment in order to support the agreements that were made under the Clean &
Energy Efficient programme in The Netherlands (Ministry of the Interior and
Kingdom Relations, 2011). This was used to create a diverse package of temporary
stimulant schemes that would help in kick-starting the desired market development of
energy efficient technologies.
In this experiment, the impact of this policy is measured under three possible
scenarios: 1. Subsidies only for insulation, 2. Subsidies only for solar boilers, and 3.
Subsidies for both insulation and solar boilers. Subsidies are provided as a percentage
13
of the total cost of implementing either of the two technologies. Also, the subsidy
percentages are higher for the low income groups and lower for the high income
groups: first quartile — 60%, second quartile — 40%, third quartile — 20%, and fourth
quartile — 10%. These subsidies are provided only for a 5-year period (2015-2020) in
order to see if this policy can indeed help in kick-starting market development on its
own.
As shown in Figure 18, the impact of providing subsidies on solar boilers is higher
than the impact of providing subsidies for insulation. This is due to the relatively
higher cost of insulation. Even after the provision of subsidies, very few households
can actually adopt insulation measures since this involves renovating their existing
house or completely demolishing and constructing a new one. Also, even if the
subsidies given to insulation are set higher than those given to solar boiler
installations, the diffusion towards solar boilers is found to be higher; this is because
the overall implementation cost for insulation still remains higher. Additionally, when
subsidies are provided for both insulation and solar boilers, there is not much change
in the aggregate consumption trend when compared to the change caused by solar
boiler subsidy.
Thus, the results show that the subsidies policy is highly effective in kick-starting the
desired market development for solar boilers, but not for insulation. Moreover, these
policies mainly affect the 1st and 2°4 income quartiles. For them, the perceived
attractiveness of solar boilers dramatically changes.
Total consumption
5.25 B
5B
2013 2017 2021 2025 2029 2033 2037 2041 2045 2049
Time (Year)
Total consumption ; BaseCase
Total consumption : Subsidy Insulation
Total consumption : Subsidy” Solar
Total consumption : Subsidy” solar_and_insulation
Figure 18: Aggregate gas consumption trend after implementing subsidies policy
3.4, Driving innovation and R&D
In their Plan of Action for Energy Saving in the Built Environment (Ministry of the
Interior and Kingdom Relations, 2011), the Ministry acknowledges the importance of
driving innovation and R&D in order to realize more energy savings in the long-run.
The innovation program Energiesprong [Energy Leap] was taken up to stimulate these
innovations. This aims for a reduction of 50% energy consumption in the built
environment by 2030 (in comparison to 1990).
In this experiment, we examine the impact of a change in learning rates for both cost
and efficiency of solar boilers and insulation caused through innovations and R&D.
14
First, the leaming rates of both the implementation costs are subjected to sensitivity
testing around the baseline value:
1. For learning rate of Insulation cost — Baseline: 0.6, Sensitivity: 0.3- 0.9.
2. For learning rate of solar boiler cost — Baseline: 0.8, Sensitivity: 0.6- 0.9.
The results of these sensitivity runs are shown in Figures 19 and 20. As can be seen,
only a decrease in the cost of insulation will lead to a small decrease in energy
consumption through higher adoption of insulation measures. However, there is no
impact on the adoption of solar boilers even if the cost of solar boiler falls. This is
because the cost of solar boilers is already very low.
The learning rates for efficiency of solar boilers and insulation are also subjected to
sensitivity testing around the baseline value:
1. For learning rate of Insulation efficiency — Baseline: 0.6, Sensitivity: 0.3- 0.9.
2. For learning rate of solar boiler efficiency — Baseline: 0.8, Sensitivity: 0.6 -
0.9.
The results of these sensitivity runs are shown in Figures 21 and 22. As can be seen,
there is great potential for energy saving by causing an increase in efficiency of either
or both technologies. Hence, while driving innovation, research that results in higher
technical efficiency should be given preference to research that focusses on
optimizing the cost of either technology in orderto make it more accessible.
Innovation insulation cost Innovation solar cost
50% = 75% (NN 95% 9 100% ay 50% 75% (ND 95% BN 100% Ba
‘lsomemion Total consumption
6B
5.75B 515B
55B 55B
5.25B 5.25B
5 Boag 2022 2032 2041 2050 5Borg Ts] DoT 2050
Time (Year) Time (Yea)
Figure 19: Sensitivity run for learning rate ofinsulation cost Figure 20: Sensitivity run for learning rate of solar cost
Innovation solar_efficiency
50% 75% (MN 95% I 100%
Total consumption
6B
5.75B
5.25B
5 Borg 2022 2032 2041 2050
Time (Year)
Figure 21: Sensitivity run for learning rate ofsolar efficiency
15
Innovation insulation efficiency
50% 75% (NIB 95% BN 100%
Total consumption
7B
5 Bog 2022 2032 2041 2050
‘Time (Year)
Figure 22: Sensitivity run for learning rate ofinsulation efficiency
3.5. Advertisement policy
The last policy that was tested for effectiveness was advertisement
promotion/awareness creation. An advertisement/awareness policy can aim at
increasing the effectiveness of advertisements or awareness programs that already
exist. This can be done by increasing the number or the range of campaigns related to
the adoption of solar boilers and insulation, or by focusing advertisement on smaller
target groups (municipality level instead of national level).
For examining the possible effects of such an investment on advertisement, a
sensitivity analysis was done for a wide range of advertisement effectiveness factor
for both technologies. The factor was varied between 0 and 0.1 (baseline being
assumed as 0.05 for both solar and insulation). Figures 23 and 24 depict the impact of
increased advertisement effectiveness on the overall gas consumption. As can be seen,
at higher advertisement rates there is considerable decrease in the aggregate gas
consumption. This happens because of an increase in mouth-to-mouth adoption. As
more households get exposed to these technologies, they not only adopt them (given
that the other financial and technical aspects remain congenial) but they also
communicate this to the households that have still not adopted. Also, as more
households adopt either of the two technologies, the learning curves for both cost and
efficiency grow at a faster rate and hence facilitate the growth again through a
positive feedback loop (Kersten et al., 2011; Malyusz & Pém, 2013).
Interestingly, on implementation of this policy, the higher income quartiles have a
higher rate of adoption than the lower income quartiles. This is conceivable due to the
fact that lower income households are restricted by financial constraints even after
being exposed to the use/impact of these two technologies. This variation in the rate
of adoption by different income groups can be seen in Figures 25 and 26. Figure 25
shows the change in composition of the number of households with only solar boilers
when the advertisement effectiveness of solar boilers is increased from 0.05 to 0.1.
Figure 26 shows the change in composition of the number of households with only
insulation when the advertisement effectiveness of insulation is increased from 0.05 to
0.1.
Hence, investing in advertisements or at least facilitating the awareness process will
have a huge impact on the overall energy trend in built environment. However, this
may not be a socially inclusive policy unless the lower income groups are being
supported with subsidies also.
16
housshold
Advertisement_policy Solar_sensitivity
50% = 75% 95% I 100%
Total consumption
7B
65B
|
5.5B
5 Bo 2022 2032 2041 2050
Time (Year)
Figure 23: Sensitivity on advertisement effectiveness of solar
Advertisement _policy_insulation_sensi!
50% 75% MN 95% BH 100%
Total consumption
7B
6.5B
68 _—_——
5.5B
5 Bo 2022 2032 2041 2050
Time (Year)
Figure 24: Sensitivity on advertisement effectiveness ofinsulation
Households with only Insulation
200,000
150,000
100,000
50,000 CPEPR PERE SSE ERE EET
0
2013 2017 2021 2025 2029 2033 2037 2041 2045 2049
Time (Year)
only firstquartile] : BaseCase
only dependent, firstquartile) : policy insulation
Households with only depend + BaseCase
only depend : _policy_insulation
Households with only dependent, thirdquartile] : BaseCase
only dependent, thirdquartile} : A dv policy_insulation
only ile] : BaseCase
only den policy_insulation
Figure 25: Impact of advertisement policy for only insulation (based on income groups)
17
Households with only Solar boilers
10,000 Ninypaeee= == ==="
2013 2017 2021 2025 2029 2033 2037 2041 2045 2049
Time (Year)
Households with only Sok firstquartile] : BaseCase
Households with only Sole firstquartile) - policy Solar
BaseCase
Households with only Sole policy Solar
Households with only Sole thirdquartile] : BaseCase
Households with only So thirdquartile] _ policy Solar
ly BaseCase
Households with only Sol policy Solar
Figure 26: Impact of advertisement policy for only solar boilers (based on income groups)
4. Discussion and C onclusions
While several researchers have already studied the impact of regulations adopted
under the European guideline for energy performance of buildings (EPBD), a
systemic study of the impact of policies related to the provision of financial stimulants,
driving of innovation, demolition of inefficient buildings, and creation of awareness
about energy efficient technologies, is lacking. Specifically, with respect to two of the
most important technologies that have great potential to drive energy saving — namely,
Insulation and Solar boilers. So, for this purpose a System Dynamics model was
developed in this study in order to explore the impact of these four policies on the
adoption of either/both of these technologies, and the result of this adoption on the
aggregate gas consumption of the Built environmentin Netherlands.
The model includes four different types of households — ones without
insulation/solar, ones with insulation but no solar, ones with solar but no insulation,
and the ones with both insulation and solar. Within each type, the houses are
classified based on their technical characteristics — independent houses, row/terraced
houses, and apartments. Also, in order to capture the social complexity involved in
decision making, the households are divided into four income quartiles under each of
the above classifications.
In the first experiment, a hypothetical situation about the future was simulated by
presuming that everything will go on as it is. This experiment revealed that the
aggregate gas consumption in the residential sector is diminishing without the
intervention of any policy. However, the rate of decrease is not too high and hence
subsequent experiments were conducted to see if this rate could be facilitated through
the four policies mentioned above.
In the second experiment, the construction of dwellings with insulation is
promoted by doubling the demolition rate of non-insulated buildings over a period of
20 years starting from 2015. It was seen that this policy helps in setting the trend fora
long-term impact on the reduction of aggregate gas consumption. Hence, setting
18
certain minimum allowable insulation standards will certainly facilitate the energy
transitions process in the Built environment.
In the third experiment, subsidies were provided for the implementation of
insulation and/or solar boilers. It was seen that the diffusion towards solar boiler
adoption is significantly higher when compared to insulation adoption. Hence,
provision of subsidies for solar boilers will prove more effective to cause a greater
decrease in the aggregate gas consumption.
In the fourth experiment, the impact of investing in innovation and R&D was
studied by observing the system’s response to less costly and more efficient solar
boilers and insulation measures. It was found that researches that result in higher
technical efficiency are more effective than researches that focus on optimizing the
cost.
In the last experiment, the effectiveness of investments in advertising or the
facilitation of awareness process is studied. It was found that increased advertisements
indeed have a huge impact on the overall energy transitions process. However, the
transitions are greater among the higher income quartiles when compared to the lower
income quartiles. Hence, this policy could be made socially inclusive in combination
with the subsidies policy.
The model used for this study has certain limitations and drawbacks as any other
model. Although several micro-level aspects related to the social behavior of
households were modelled, it would be appropriate to say that the chosen level of
aggregation is rather high. The Dutch dwelling stock was grouped into three
homogenous types, thus overlooking the heterogeneity that exists in reality. However,
this structural simplicity was needed in order to capture the long-term dynamics at a
higher aggregation level. The uncertainty around different parameters that were
assumed during the modelling process is also an important issue. To account for this
uncertainty, and to ensure robustness in behavior, sensitivity analyses were performed
on all the important and uncertain parameters. Although the results are found to vary
numerically, the general behavior of all key performance indicators are found to stay
intact overa wide range of values in these uncertain parameters.
Lastly, the scope of this study could be expanded in many ways. The inclusion of
rebound effect while modelling consumer behavior, the inclusion of gas consumption
by households for cooking and other purposes, incorporation of non-energy related
expenses of households, and the provision of energy tax based on energy labels or
EPC values could be some aspects that would make the system boundaries more
realistic. Furthermore, different technical characteristics of dwellings, such as
building design and surface area, can be used to impact the diffusion process. Market
limitations posed by the lack of ability to cope with a sudden increase in demand for
either of the two technologies could also be incorporated. Also, other energy efficient
measures related to the consumption of natural gas could be included in the study.
Finally, with relevant alterations the model can be extended to include the industrial
and commercial sector.
REFERENCES
Beerepoot, M. 2007. Public energy performance policy and the effect on diffusion of
solar thermal systems in buildings: A Dutch experience. Renewable Energy 32: 1882-
1897.
19
Ben Maalla, E.M., Kunsch, P.L. 2008. Simulation of micro-CHP diffusion by means
of System Dynamics. Energy Policy 36: 2308-2319.
Bogerd, A., van Oel, C., de Haas, G. 2009. Occupant-owners preferences in deciding
about innovative renovation concepts in changing Housing Markets: Integration and
Segmentation, In Proceedings of ENHR 2009 Conference, Prague.
Buildings Performance Institute Europe (BPIE). 2010. Energy Performance
Certificates across Europe: From design to implementation. Brussels, Belgium.
Council of the European Union. 2007. 7224/1/07 REV 1 Presidency Conclusions,
Brussels.
Fallah-Fini, S., Rahmandad, H., Chen, HJ., Wang, Y. 2013. Connecting micro
dynamics and population distributions in system dynamics models. System Dynamics
Review 29(4): 197-215.
Grosser S., Ulli-Beer S., Mojtahedzadeh M. 2006. Diffusion Dynamics of Energy-
Efficient Innovations in the Residential Building Environment. In Proceedings of the
24th International Conference, Nijmegen, Netherlands, System Dynamics Society.
ING. 2013. Saving Energy in the Netherlands: New Directive takes on Energy
Efficiency. Amsterdam, Netherlands.
Kersten F., Doll, R., Kux, A., Huljic, D.M. 2011. PV Leaming curves: Past and
Future Drivers of cost reduction. In Proceeding of the 26th European Photovoltaic
Solar Energy Conference, Hamburg.
Malyusz, L., Pém, A. 2013. Prediction of the leaning curve in roof insulation.
Automation in Construction 36:191-195.
Ministry of the Interior and Kingdom Relations. 2011. Plan of Action: Energy Saving
in Built Environment, Netherlands. Retrieved November 20, 2013, from
http://www.government.nl/documents-and-publications/reports/2011/02/25/plan-of-
action-energy-saving-in-built-environment.html,
Sterman, J.D. 2000. Business Dynamics: Systems Thinking and Modeling for a
Complex World. Irwin/McGraw-Hill, Boston.
Yucel, G. 2013. Extent of inertia caused by the existing building stock against an
energy transition in the Netherlands. Energy and Buildings 56: 134-145.
Van Raaij, W.F., Verhallen, M.M. 1986. Determinants of Home Insulation Intention.
Advances in Consumer Research, UT: Association for Consumer Research, 13: 476-
480
World Development Indicators database, W orld Bank. Accessed April 23, 2014, from
http://data.worldbank. orq/data-catalog/world-development-indicators.
20
2.4 VULNERABILITIES’ PATHWAYS OF THE HOTSPOT
The integrated structure presented in Error! Reference source not found. allows us not only to identify what
shocks and stresses the food system at stake, but also to systemically explore how the disturbances are conveyed
throughout key feedback processes in the food system and generate vulnerabilities.
2.4.1 Vulnerability pathway I: Dependence on external inputs
The strong reinforcing feedback loops that drive food production through intensification (R1b, Figure 4) and
mechanization (R1a, Figure 4) and concurrently degrade natural resources and erode tacit knowledge, give rise
also to two additional strong reinforcing processes through which degraded natural resources are compensated
for with external inputs (R2, Figure 5) and tacit k ledge is it by i (R4, Figure
6). Both of the latter reinforcing feedback loops are examples of unintended processes that increasingly lock food
producers into dependence on external inputs, the companies that provide them and the capitalist relationships
of food production that frame their decisions [5,108]. The use of external inputs considerably changes food
producing practices as well as agroecosystems in which they are applied, giving rise to unintended consequences
(e.g., weed resistance, pollinator decline) that are then stabilized with new external inputs (e.g., stronger
herbicide cocktails) that in turn end up reinforcing the dependency. The result is that food production is based
on continuous reir in engi rather than tacit knowledge and ecosystems resilience
(condition of natural resources). Therefore, if for some reasons (e.g., fossil fuels scarcity, geopolitical tensions,
economic crisis) external inputs were not available for food producers: first, it will take a long time for an
alternative food production paradigm to become effective (because of, for instance, the need to rebuild the
stocks of tacit knowledge and natural resources condition) and second, the outcomes could potentially be far
more undesirable than that of a system which never used those stabilizers. Moreover, relying on a limited range
of ‘stabilizing’ external inputs makes the food system particularly vulnerable to disturbances that operate beyond
their scope of fixes such as unexpected and non-linear climate change and feedbacks.
2.4.2 Vulnerability pathway II: Striving for efficiency while losing r
The conventional European food system manages its growth and expansion based on ideas of maximizing
efficiency realized through inter alia scale i jalization and ical innovation (i.e., balancing
loop of cost maximization (B8, Figure 8) that perpetuates a strong reinforcing feedback loop of efficiency
maximization (R5, Figure 8)). Food producers across Europe experience effects of the cost minimization processes
in many different ways. Scale economies force many small- and medium-scale food producers out of the agri-
food business entirely, which is evident in the declining number of farms. This trend along with the strong
reinforcing spiral of labor reduction (Ric, Figure 4) translates into increasingly fewer people in society with
knowledge and skills to produce food, implying further decline in the stock of tacit knowledge (Figure 6) as well
as disruption in rural communities, both of which have been found crucial for resilience of food systems to shocks
and stresses [14,21,109,110]. Besides, scale economies drive consolidation (i.e., growing farm size), and hence
reduce the diversity of scale at which food producers operate. Specialization is apparent, for instance, in the
trend towards a single dominant activity on farms and widespread monocultures. Currently, in the EU almost
half of the holdings are specialized in cropping and 27% in livestock [111]. Accordingly, as the system specializes,
the diversity of organizational forms as well as crops and animals decreases in the food system. Technical
innovations (e.g., application of more and more specific fertilizers, herbicides and pesticides and genetic
advances) to a great extent are in hands of few multinational corporations [5]. This narrows down sources of
technical innovations as well as the range of choices of ‘what’ and ‘how’ to produce that food producers have.
For instance, commercial seeds and breeds focus on a few traits in a few crops, forcing food producers to base
their production on these traits. The three processes seem to favor each other, so that, for instance, the technical
innovations (e.g., promotion of agrochemical use, biotechnology, single crop machinery, etc.) are most (costs)
through scale ies [46] and ialization [5], [9], [99]. Common feature of all of these processes
is that they increase efficiency of food production, but at the same time decrease diversity of different elements
in the system. The latter is, in turn, crucial for absorption of shocks and stresses, adaptation and alternative
solutions [9], [104]-[106]. Having low diversity in the food system allows disturbances to become augmented,
both economically (e.g., food pricing controlled by few) and ecologically (e.g., contamination on a single farm
can easily effect the entire country). Thus, it seems that through strong efficiency maximization loop (R5, Figure
8) food producers trade-off short-term productivity against long-term resilience.
In essence, vulnerabilities in the conventional European food system arise if disturbances strengthen the
reinforcing feedback loops and further weaken or delays the balancing loops. For instance, climate change
related shocks such as drought, flood or storm, will likely strengthen the intensification reinforcing feedback loop
(R1b, Figure 4) because of yield losses. Yield losses increase the pressure on food producers to produce more,
disregarding the balancing loops of natural resources degradation and regeneration (B1, B2 Figure 5), thus
further lowering the stock of natural resources condition. When the stock depletes, yield declines, and translates
into undesirable outcome of reduced food production and hence food insecurity.
2.5 ORGANIC FARMING AS POLICY TO ADDRESS THE HOTSPOT’S VULNERABILITIES
Based on the analysis in previous sections, we argue that the European food system based on conventional
agriculture is vulnerable. An alternative approach to food system, which does not trade-off long-term resilience
for productivity and stability, is called for [5,7—9]. King [14] lists several potential approaches for a resilient food
system, including organic and biodynamic farming, permaculture, farmers’ markets, community-supported
agriculture and community gardens. In Europe organic farming is the fastest growing of all alternatives to the
conventional food system, which is regulated at EU level and receives considerable public financial support.
However, is transition to organic farming a viable policy for making the European food system more resilient?
2.5.1 Resilience pathway |: Low external input system
2.5.1.1 Potential
Organic farming as per definition is a low external input system with inter alia diversification and nutrient
cycling at its heart. As mentioned in 2.5.1.1 and 2.5.2.1 it preserves higher stocks of natural resources and tacit
knowledge as well as has better recognized and operating balancing loops (B1, B2, Figure 5; B3, Figure 6), food
producers may thus escape from being locked into the dangerous dependence on external inputs (R2, Figure 5 ;
R4, Figure 6).
2.5.1.2 Limitation
However, implementation of organic food production principles in practice is diverse and ranges from mere
‘input substitution’ to fundamental ‘system redesign’ [120]. This implies that there are organic food producers,
of which practices diverge only slightly from conventional practices [27]. As organic food producers are not
rewarded for continuous improvement, but have to comply just with minimum standards, they are incentivized
to simply substitute prohibited with allowed inputs sourced from outside of the system. As a result they will be
again locked into the vicious circles creating dependence on external inputs (R2, Figure 5 ; R4, Figure 6) with all
its consequences for resilience of the prevailing food system.
2.5.2 Resi
nce pathway II: Striving for diversification
2.5.2.1 Potential
In addition to better environmental outcomes, many studies have found that organic food producers
perform better also in socio-economic terms as compared to their conventional counterparts [112]. Simply
looking at comparisons of organic versus conventional short-term profitability, organic seems to be a promising
option to preserve viability of farms. Besides, organic food system is characterized also by diversity of markets
(e.g., specialized organic food stores, farmers’ markets and direct farm marketing, food baskets), through which
organic food is provided to consumers. These two features — better financial performance and diversity of
markets — suggest that potentially the internal market structure of the organic food system is different from the
conventional one and that the system can address the vulnerabilities related to socio-economic organization of
food production inherent in the latter.
2.5.2.2 Limitation
However, there are many signs indicating that organic food system based on certification of food production
methods alone, falls into the same reinforcing mechanisms as conventional system and gives up its resilient
features for efficiency (R5, Figure 8) and itself is vulnerable. For instance, establishing certification put barriers
for smaller food producers to enter the sector, because of costs and because it facilitates larger retailers to sell
organic products [9]. Hence, the organic food system becomes more and more consolidated and losses its
diversity, which has consequences for contributing to resilience of the conventional European food system.
3 CONCLUSIONS
In this paper, we have proposed a new way to help policymakers understand the food system’s
vulnerabilities and assess whether organic farming can enhance its resilience. For this, we adopted a system
dynamics approach to capture the dynamic complexity of the food system. We have identified a number of key
systemic vulnerabilities, including the degradation of the natural resource base of food production, the erosion
of its knowledge base, its dependence on external inputs, the latent instability on agri-food markets and the
strive for efficiency. We have argued that organic farming has the potential to address these vulnerabilities, but
at the same time risks of falling into the same systemic pitfalls through a process of conventionalization. More
specifically, organic farming as a food system has to be carefully designed and implemented to overcome the
contradictions between the dominant socio-economic organization of food production and the ability to
implement holistic understanding of organic principles on a broader scale. Further research needs to identify
policy interventions that allow organic farming to reach its full potential avoiding these pitfalls.
4 ACKNOWLEDGEMENTS
This paper originates from an EU FP7 funded project TRANSMANGO “Assessment of the impact of global
drivers of change on Europe's food security”; Grant agreement no: 613532; Theme KBBE.2013.2.5-01. Andreas
Gerber is supported by the Norwegian Research Council through the project “Simulation based tools for linking
knowledge with action to improve and maintain food security in Africa” (contract number 217931/F10).
5 REFERENCES
1. Marten, G. G.; Atalan-Helicke, N. Introduction to the Symposium on American Food Resilience. J. Environ. Stud. Sci. 2015,
5, 308-320.
2. Hazell, P.; Wood, S. Drivers of change in global agriculture. Philos. Trans. R. Soc. B Biol. Sci. 2008, 363, 495-515.
3. Household Consumption Expenditure - National Accounts. Available online: http://ec.europa.eu/eurostat/statistics-
lained/index.ph hold_« i »_-_national_accounts (accessed on 13 April 2016).
4. Kirchmann, H.; Thorvaldsson, G. Challenging targets for future agriculture. Eur. J. Agron. 2000, 12, 145-161.
5, Hendrickson, M. K. Resilience in a concentrated and consolidated food system. J. Environ. Stud. Sci. 2015.
6. Tansey, G. Food and thriving people: paradigm shifts for fair and sustainable food systems. Food Energy Secur. 2013, 2, 1-
11,
7. International Assessment of Agricultural Knowledge, Science and Technology for Development. Agriculture
at a Crossroads - Global Report 2009. Available online:
http; .unep.org/di ports/IAASTD/EN/Agric 620at%20a%20Crossroads_Global%20Report%2
O(English) pdf (accessed on 1 February 2016).
8, The 3rd SCAR Foresight Exercise. Sustainable Food Consumption and Production in a Resource-Constrained World 2009.
Available online: https://ec.europa.eu/research/agriculture/scar/pdf/scar_feg3_final_report_01_02_2011.pdf (accessed on
1 February 2016).
9. Rotz, S.; Fraser, E. D. G. Resilience and the industrial food system: analyzing the impacts of agricultural industrialization on
food system vulnerability. J. Environ. Stud. Sci. 2015.
10. Godfray, H.C. J.; Crute, |. R.; Haddad, L.; Lawrence, D.; Muir, J. F.; Nisbett, N.; Pretty, J.; Robinson, S.; Toulmin, C.; Whiteley,
R. The future of the global food system. Philos. Trans. R. Soc. Lond. B. Biol. Sci. 2010, 365, 2769-77.
11. Sundkvist, A.; Milestad, R.; Jansson, A. On the importance of tightening feedback loops for sustainable development of
food systems. Food Policy 2005, 30, 224-239.
12. Ingram, J. A food systems approach to researching food security and its interactions with global environmental change.
Food Secur. 2011, 3, 417-431.
13. Stave, K.; Kopainsky, B. A system dynamics approach for examining mechanisms and pathways of food supply
vulnerability. J. Environ. Stud. Sci. 2015.
14. King, C. A. C ity resilience and y agri-ecological systems: reconnecting people and food, and people
with people. Syst. Res. Behav. Sci. 2008, 25, 111-124.
15. Stolze, M.; Lampkin, N. Policy for organic farming: Rationale and concepts. Food Policy 2009, 34, 237-244.
16. FiBL & IFOAM. The World of Organic Agriculture. Statistics & Emerging Trends 2016. Available online:
https://shop.fibl. i hop/169: ‘ganic-world-2016.pdf (accessed on 1 February 2016).
17. Niggli, U. Sustainability of organic food production: challenges and innovations. Proc. Nutr. Soc. U. S. A. 2015, 74, 83-8.
18. Darnhofer, |. Contributing to a transition to sustainability of agri-food systems: potentials and pitfalls
for organic farming. In Organic Farming, Prototype for Sustainable Agricultures; Bellon, S., Penvern, S., Eds.,; Springer:
Dordrecht Heidelberg New York London, 2014; pp. 439-452.
19. Reganold, J. P.; Wachter, J. M. Organic agriculture in the twenty-first century. Nat. Plants 2016, 2, 15221.
20. Food Security Information Network. Resilience Measurement Principles 2014. Available __ online:
http: fao.ore/fi i _upload/droug FSINY ii 20 0201401.pdf (accessed
on 20 March 2016).
21. Milestad, R.; Darnhofer, |. Building farm resilience: the prospects and challenges of organic farming. J. Sustain. Agric. 2003,
22, 81-97.
22. Food and Agriculture Organization. Building Resilience for an Unpredictable Future: How Organic Agriculture Can Help
Farmers Adapt to Climate Change 2006. Available online: http://www.fao.org/3/a-ah617e.pdf. (accessed on 30 March 2016).
23. Darnhofer, |. Strategies of family farms to strengthen their resilience. Environ. Policy Gov. 2010, 20, 212-222.
24. Scialabba, N. E.-H.; Miller-Lindenlauf, M. Organic agriculture and climate change. Renew. Agric. Food Syst. 2010, 25, 158-
169.
25. LUPG. The Role of Agroecology in Sustainable Intensification 2015. Available online:
http://www.snh.gov.uk/docs/A1652615.pdf (accessed on 1 February 2016).
26. Guthman, J. Agrarian Dreams: The Paradox of Organic Farming in California. University of California Press: Berkeley, Los
Angeles, London, 2004.
27. Ponti, T. De; Rijk, B.; Ittersum, M. K. Van The crop yield gap between organic and conventional agriculture. Agric. Syst.
2012, 108, 1-9.
28. Crowder, D. W.; Reganold, J. P. Financial competitiveness of organic agriculture on a global scale. Proc. Natl. Acad. Sci. U.
S.A. 2015, 1-6.
29. Gomiero, T.; Pimentel, D.; Paoletti, M. G. Environmental Impact of Different Agricultural Management Practices:
Conventional vs. Organic Agriculture. CRC. Crit. Rev. Plant Sci. 2011, 30, 95-124.
30. Palupi, E.; Jayanegara, A.; Ploeger, A.; Kahl, J. Comparison of nutritional quality between conventional and organic dairy
products: A meta-analysis. J. Sci. Food Agric. 2012, 92, 2774-2781.
31. European Commission. Farm Economics Briefs. Organic versus conventional farming, which performs better financially?
Available online: http://ec.europa.eu/agriculture/rica/pdf/FEB4_Organic_farming_final_web.pdf (accessed on 10 March
2016).
32. Birkhofer, K.; Bezemer, T. M.; Bloem, J.; Bonkowski, M.; Christensen, S.; Dubois, D.; Ekelund, F.; Flie8bach, A.; Gunst, L.;
Hedlund, K.; Mader, P.; Mikola, J.; Robin, C.; Setala, H.; Tatin-Froux, F.; Van der Putten, W. H.; Scheu, S. Long-term organic
farming fosters below and aboveground biota: Implications for soil quality, biological control and productivity. Soil Biol.
Biochem. 2008, 40, 2297-2308.
33. Bengtsson, J.; Ahnstrém, J.; Weibull, A.-C. The effects of organic agriculture on biodiversity and abundance: a meta-
analysis. J. Appl. Ecol. 2005, 42, 261-269.
34. Tuomisto, H. L.; Hodge, |. D.; Riordan, P.; Macdonald, D. W. Does organic farming reduce environmental impacts? — A
meta-analysis of European research. J. Environ. Manage. 2012, 112, 309-320.
35. Ericksen, P. J. Conceptualizing food systems for global environmental change research. Glob. Environ. Chang. 2008, 18,
234-245.
36. Ericksen, P. J. What is the vulnerability of a food system to global environmental change? 2008, 13(2), 14.
37. Hammond, R. a; Dubé, L. A systems science perspective and transdisciplinary models for food and nutrition security. Proc.
Natl. Acad. Sci. U.S. A. 2012, 109, 12356-63.
38. Sterman, J. Business Dynamics: Systems Thinking and Modeling for a Complex World. McGraw-Hill Higher Education:
Toronto, 2000.
39. Tendall, D. M.; Joerin, J.; Kopainsky, B.; Edwards, P.; Shreck, A.; Le, Q. B.; Kruetli, P.; Grant, M.; Six, J. Food system
resilience: Defining the concept. Glob. Food Sec. 2015, 6, 17-23.
40. Ingram, J.; Ericksen, P.; Liverman, D. Food Security and Global Environmental Change, Earthscan: London, Washington,
DC, 2010.
41. Quinlan, A. E.; Berbés-Blézquez, M.; Haider, L. J.; Peterson, G. D. Measuring and assessing resilience: Broadening
understanding through multiple disciplinary perspectives. J. Appl. Ecol. 2015, 53, 677-687.
42. Ford, A. Modeling the Environment, 2nd ed.; IslandPress, 2009.
43. Ford, J. D. Vulnerability of Inuit food systems to food insecurity as a consequence of climate change: A case study from
Igloolik, Nunavut. Reg. Environ. Chang. 2009, 9, 83-100.
44, Hoffmann, M. H. G. Cognitive Conditions of Diagrammatic Reasoning. Georgia Tech's School of Public Policy Working
Paper Series 2007, 24 Available online: http://works.bepress.com/michael_hoffmann/1/ (accessed on 20 March 2016).
45. A Sustainability Institute Report 2003. C ity System Challenges Moving inability into the Mai of
Natural Resource Economies. Available online:
http://s3. c |_storage/w i ies.com/ContentPages/707629742.pdf (accessed on 12
December 2015).
46. Varian, H. R., Intermediate Microeconomics, 8th ed.; WW Norton & Co: New York, London, 2010.
47. Rasmussen, S. Production Economics: The Basic Theory of Production Optimisation, 2nd ed.; Springer: Heidelberg New
York Dordrecht London, 2011.
48. Debertin, D. L. Agricultura Production ics, Ind ed.; i ishing Company: New York, USA, 2012.
50. European Commission, 50 years of food safety in the EU 2007. Available _ online:
http://ec.europa.eu/food/food/docs/SOyears_foodsafety_en.pdf (accessed on 20 April 2016).
51. Svato8, M. Selected trends forming European agriculture. Agric. Econ. 2008, 54, 93-101.
52. FAO. World Outlook and State of Food and Agriculture - 1950. Available _ online:
http://www.fao.org/docrep/016/ap638e/ap638e.pdf (accessed on 1 February 1950).
53. Godfray, H.C. J.; Beddington, J. R.; Crute, |. R.; Haddad, L.; Lawrence, D.; Muir, J. F.; Pretty, J.; Robinson, S.; Thomas, S. M.;
Toulmin, C. Food Security: The Challenge of Feeding 9 Billion People. Science 2012, 327, 5967, 812-818.
54. Economist Intelligence Unit (EIU). 2014. Food security in focus: Europe 2014. Avaliable online:
http://foodsecurityi eiu.com/H D fil U%20GFSI%202014_Europe%20regional%20repo
rt.pdf (accessed on 20 March 2016).
55. Swinnen, J. F. M.; Banerjee, A. N.; De Gorter, H. Economic development, institutional change, and the political economy
of agricultural protection An econometric study of Belgium since the 19th century. Agric. Econ. 2001, 26, 25-43.
56. EU Agricultural Outlook. Prospects for EU agricultural markets and income 2015-2025. Avaitable online:
http://ec.europa.eu/agriculture/ d-p t itlook/2015/fullrep_en.pdf (accessed on 1 February
2016).
57. de Vries, B. J. M. Sustainability Science; Cambridge University Press: Cambridge, 2013.
58. Brisson, N.; Gate, P.; Gouache, D.; Charmet, G.; Oury, F. X.; Huard, F. Why are wheat yields stagnating in Europe? A
comprehensive data analysis for France. F. Crop. Res. 2010, 119, 201-212.
59. Grassini, P.; Eskridge, K. M.; Cassman, K. G. Distinguishing between yield advances and yield plateaus in historical crop
production trends. Nat. Commun. 2013, 4, 2918.
60. European Commission. EU Agricultural Markets Briefs. Price developments and links to food security - price level and
volatility. Available online: http://ec.europa.eu/agriculture/markets-and-prices/market-briefs/pdf/05_en.pdf (accessed on
10 March 2016).
61. Communication from the Commission to the European Parliament, the Council, the European Economic and Social
Committee and the Committee of the Regions - Food Prices in Europe. COM(2008) 821. Available online: http://eur-
lex.europa.eu/legal-content/EN/TXT/PDE/?uri=CELEX:52008DC0821&from=EN (accessed on 1 February 2016)
62. Loopstra, R.; Reeves, A.; Stuckler, D. Rising food insecurity in Europe. Lancet 2015, 385, 2041.
63. WHO. Obesity Europe. Available online: http://www.euro.who.int/en/health-topic: icable-di besity
(accessed on 20 March 2016).
64. European Commission. EC Farm Economics Briefs. Income Developments in EU Farms. Available online:
http://ec.europa.eu/agriculture/rica/pdf/Brief201101.pdf (accessed on 10 March 2016).
65. European Commission. EC Farm Economics Briefs. EU Production Costs Overview. Available online:
http://ec.europa.eu/agriculture/rica/pdf/Brief201102. pdf (accessed on 10 March 2016).
66. European Commission. EU Agricultural Economics Briefs. Structure and Synamics of EU Farms: Changes,
Trends and Policy Relevance. Available online: http://ec.europa.eu/agriculture/rural-area-economics/briefs/pdf/09_en.pdf
(accessed on 10 March 2016).
67. European Commission. EU Agricultural Economics Briefs. EU Agricultural Income 2014 — First Estimates. Available online:
http://ec.europa.eu/agriculture/rural-area-economics/briefs/pdf/003_en.pdf (accessed on 10 March 2016).
68. Matthews, A. FADN data highlights dependence of EU farms on subsidy payments. Available online:
http://capreform.eu/fadn-data-highlight: fi bsidy (accessed on 10 March 2016).
69. Enjolras, G.; Capitanio, F.; Aubert, M.; Adinolfi, F. Direct payments, crop insurance and the volatility of farm income. Some
evidence in France and in Italy. New medit Mediterr. J. Econ. Agric. Environ.2014, 13, 31-40.
70. Rabobank. Rethinking the F&A Supply Chain. Impact of Agricultural Price Volatility on Sourcing Strategies. Available online:
http://www.boerderij.n|/pagefiles/35979/002_boerderij-download-agd573390d01.pdf (accessed on 10 March 2016).
71. EEA. Agriculture and Environment in EU-15 — The IRENA Indicator Report 2005. Available online:
file:///C:/Users/u0095636/Downloads/EEA_report_6_2005.pdf (accessed on 20 March 2016).
73. Rockstrom, J. et al. Planetary boundaries: Exploring the safe operating space for humanity. Ecol. Soc. 2009, 14.
74, EEA Food Security and Environmenta Impacts. Available online:
http: .eea.europa. eri igreening-agric policy/food-security-and. impacts
(accessed on).
75. Kopainsky, B.; Huber, R.; Pedercini, M. Food provision and environmental goals in the swiss agri-food system: System
dynamics and the social-ecological systems framework. Syst. Res. Behav. Sci. 2015, 432, 414-432.
76. Olper, A.; Raimondi, V. Patterns and determinants of off-farm migration transfer frictions and persistency of relative
income gaps. Ceps 2013, 32.
77. Trobe, H. L. La; Acott, T. G.; Trobd, H. L. La; Acotp, T. G. Localising the global food system. Int J Sust Dev World 2000, 74,
1350-4509.
78. Cordell, D.; Neset, T.-S. S. Phosphorus vulnerability: A qualitative framework for assessing the vulnerability of national
and regional food systems to the multi-dimensional stressors of phosphorus scarcity. Glob. Environ. Chang. 2014, 24, 108-
122.
79. Neset, T.-S. S.; Cordell, D. Global phosphorus scarcity: identifying synergies for a sustainable future. J. Sci. Food Agric.
2012, 92, 2-6.
80. Cordell, D.; Drangert, J.-O.; White, S. The story of phosphorus: Global food security and food for thought. Glob. Environ.
Chang. 2009, 19, 292-305.
81. Wallgren, C.; Mattias, H. Eating energy — Identifying possibilities for reduced energy use in the future food supply system.
Energy Pol. 2009, 37, 5803-5813.
82. Pfeiffer, D.A. Eating Fossil Fuels; The Wilderness Publications 2003.
83. Stoate, C.; Baldi, A.; Beja, P.; Boatman, N. D.; Herzon, |.; Van Doorn, A.; De Snoo, G. R.; Rakosy, L.; Ramwell, C. Ecological
impacts of early 21st century agricultural change in Europe — A review. J. Environ. Manage. 2009, 91, 22-46.
84. Olesen, J. E.; Schelde, K.; Weiske, A.; Weisbjerg, M. R.; Asman, W. A. H.; Djurhuus, J. Modelling greenhouse gas emissions
from European conventional and organic dairy farms. Agric. Ecosyst. Environ. 2006, 112, 207-220.
85. The 1st SCAR Foresight Exercise. Agriculture and Environment 2006. Available online:
https://ec.europa.eu/research/agriculture/scar/pdf/scar_foresight_environment_en.pdf (accessed on 1 February 2016).
86. ESF/COST Forward Look on European Food Systems in a Changing World 2009. Available online:
http://www.esf.org/fileadmin/Public_documents/Publications/food.pdf (accessed on 1 February 2016).
87. Webb, J. et al. Food security and nutrition and sustainable agriculture. Food Policy 2011, 149, 193-208.
88. Tilman, D.; Cassman, K.G.; Matson, P.A.; Naylor, R.; Polasky, S. Agriculture sustainability and intensive production
practices. Nature 2002. 418, 671-677.
89. Ramirez, C. A.; Worrell, E. Feeding fossil fuels to the soil An analysis of energy embedded and technological learning in
the fertilizer industry. Worrell / Resour. Conserv. Recycl. 2006, 46, 75-93.
90. Shcherbak, |.; Millar, N.; Robertson, G. P. Global metaanalysis of the nonlinear response of soil nitrous oxide (N20)
emissions to fertilizer nitrogen. Proc. Natl. Acad. Sci. 2014, 111, 9199-9204.
91. Geiger, F. et al. Persistent negative effects of pesticides on biodiversity and biological control potential on European
farmland. Basic Appl. Ecol. 2010, 11, 97-105.
92. Mclaughlin, A.; Mineau, P. The impact of agricultural practices on biodiversity. Agric. Ecosyst. Environ. 1995, 55, 201-212.
93. FAO. _Livestock's Long Shadow Environmental Issues and = Options. Available online:
ftp://ftp.fao.org/docrep/fao/010/a0701e/a0701e.pdf (accessed on 20 March 2016).
94. Tamminga, S. Pollution due to nutrient losses and its control in European animal production. Livest. Prod. Sci. 2003, 84,
101-111.
95. Goodman, D.; Sorj, B.; Wilkinson, J. From Farming to Biotechnology: A Theory of Agro-Industrial Development. Basil
Blackwell: Oxford, 1987.
96. Lundvall, B.; Johnson, B. The learning economy. J. Industry Studies. 1994. 1, 2, 23-42
97. UNIDO. Determinants of total factor productivity: a literature review 2007. Available online:
http: .unido.org//fileadmit _media/Publicati ch_and_statistics/Branch_publications/Research_and_P
olicy/Files/Working_Papers/2007/WP022007%20-%20Determinants%200f%20total%20factor%20productivity.pdf
(accessed on 20 March 2016).
98. Morgan, K.; Murdoch, J. Organic vs. ional agriculture: power and i ion in the food chain.
Geoforum 2000, 31, 159-173.
99. Cochrane, W. W. Farm price: myth and reality. University of Minnesota Press: St. Paul, 1958.
100. Godfray, H. C. J.; Crute, |. R.; Haddad, L.; Lawrence, D.; Muir, J. F.; Nisbett, N.; Pretty, J.; Robinson,
Whiteley, R. The future of the global food system. Philos. Trans. R. Soc. B Biol. Sci. 2010, 365, 2769-2777.
101. Adger, W. N. Vulnerability. Glob. Environ. Chang. 2006, 16, 268-281.
102. O’Brien, K.; Leichenko, R.; Kelkar, U.; Venema, H.; Aandahl, G.; Tompkins, H.; Javed, A.; Bhadwal, S.; Barg, S.; Nygaard,
L.; West, J. Mapping vulnerability to multiple stressors: climate change and globalization in India. Glob. Environ. Chang. 2004,
14, 303-313.
103. Magdoff, F. Food as commodity. Mon. Rev. 2012, 63, 8, 15-22
104. EUROSTAT 2013. Agri-envi Indicator - Specialisation. Available online: http://ec.europa.eu/eurostat/statistics-
lained/index.php/Agri-envi |_indicator_-_specialisation (accessed on 15 March 2016).
105. Mondelaers, K.; Aertsens, J.; G. Van A meta-analysis of the dif in envi impacts between
organic and conventional farming. Br. Food J. 2009, 111, 1098-1119.
106. Sandhu, H. S.; Wratten, S. D.; Cullen, R. Organic agriculture and ecosystem services. Environ. Sci. Policy 2010, 13, 1-7.
107. Gattinger, A.; Muller, A.; Haeni, M.; Skinner, C.; Fliessbach, A.; Buchmann, N.; Mader, P.; Stolze, M.; Smith, P.; Scialabba,
N. E.-H.; Niggli, U. Enhanced top soil carbon stocks under organic farming. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 18226-31.
108. Morgan, K.; Murdoch, J. Organic vs. ional agriculture: power and i ion in the food chain.
Geoforum 2000, 31, 159-173.
109. Seufert, V.; Ramankutty, N.; Foley, J. a. Comparing the yields of organic and conventional agriculture. Nature 2012, 485,
229-232.
110. Smit, B.; Wandel, J. Adaptation, adaptive capacity and vulnerability. Glob. Environ. Chang. 2006, 16, 282-292.
111. Lamine, C. Transition pathways towards a robust ecologization of agriculture and the need for system redesign. Cases
from organic farming and IPM. J. Rural Stud. 2011, 27, 209-219.
112. Crowder, D. W.; Reganold, J. P. Financial competitiveness of organic agriculture on a global scale. Proc. Natl. Acad. Sci.
U.S.A. 2015, 112, 7611-7616.
113. Offermann, F.; Nieberg, H.; Zander, K. Dependency of organic farms on direct payments in selected EU member states:
Today and tomorrow. Food Policy 2009, 34, 273-279.
114. Smith, E.; Marsden, T. Exploring the “limits to growth” in UK organics: beyond the statistical image. J. Rural Stud. 2004,
20, 345-357.
