5.) DISCUSSION: A NEED FOR MORE HARMONY?
5.1. Comparative studies on EIAs in the EU
There are relatively few comparative studies or assessments of EIA
legislation or practice in the EU. This might have several reasons: Firstly, the
member states are under a strict regime that obliges them to send frequent
reports on EIA implementation and practice to the commission. The commission
collects this information and typically once every five years, a thorough report
is created that is then directed at the European Parliament, the European
Economic and Social Committee and the Committee of the Regions. These bodies
formally comment on the report. The last time such a report was published was in
2009; the relevant document is a valuable source of information for this thesis
(Ref. 23). Member states might see this reporting as a sufficient auditing tool
and refrain from engaging in further international assessments.
Secondly, assessments of EIA legislation and practice are often done nationally;
such studies, comparisons are usually done only to previous legislation of that
particular member state. Comparisons with other member states touch, if at all,
only issues regarding transboundary EIAs according to the Espoo Convention. A
good example for a national evaluation of EIAs is provided by Austria in the
form of a thorough study (Ref. 15) and one of frequent reports by the relevant
ministry to the parliament (Ref. 73).
Thirdly, a few independent studies have tried to assess EIA practice in a
comparative manner, often based on case law. By far the most extensive
systematic and EU-wide assessment of EIA implementation was made in 2009 for the
“(IMP)3” report (Ref. 4) that supplemented the 5-year report of the commission.
However, even this report concluded with the words: “The research on existing
evaluation studies have [sic] shown rather poor results. Except for an
evaluation of the performance of the EIA process from 1996, only a few other
studies could be found, […]”. In the following paragraph, I will give a very
brief overview on the key-findings of those studies that I identified as
The (IMP)3 report came up with three main issues that, according to the report,
would deserve some closer attention and consideration for tackling them on the
level of EU legislation: (1) To think about a formalisation and possible
exclusion of health aspects from the EIA, so that a health impact assessment
might develop into a separate proceeding supplementing a conventional EIA; (2)
risk assessment practice might require a harmonisation; (3) screening is done
with a great deal of discretion and diverges among member states.
The 2009 report of the commission, which was supplemented by the (IMP)3 report,
also dealt with the issue of widely diverging screening practice:
“Implementation and case-law show that, when establishing thresholds, MS often
exceed their margin of discretion, either by taking account only of some
selection criteria in Annex III or by exempting some projects in advance” (Ref.
74). The report also criticises a wide-spread practice in some member states to
“salami-slice” projects to take advantage of insufficient screening of
cumulative effects. The report concludes with a clear policy recommendation:
“Thus, the screening mechanism should be simplified and clarified, for example,
by detailing the selection criteria listed in Annex III and by establishing
Community thresholds, criteria or triggers (e.g. by comitology)” (Ref. 74).
A highly interesting notion can be found in the next paragraph of the report
(3.2), when it recommends to link EIAs to environmental standards rather than
thresholds for project criteria and procedural requirements only. For example, a
power plant should not be judged based on its emissions only, but rather based
on criteria for the surrounding animal and plant communities and their reaction
to the emissions. Whilst this should already be the case and is certainly common
practice on the three countries subject to this thesis, many EU member states
appear to interpret the impacts of pollution differently. The report also
criticises the lack of documentation and auditing standards: “There are major
differences in the quality of EIA documentation, not only between different MS
but also within MS themselves”. The report then deals with the issue of EIA
duration: “The lack of provisions in the Directive relating to reasonable
timeframe and preferably fixed timeframe for granting development consent, to
the duration of the validity of the EIA and to monitoring the significant
environmental effects of the implementation of projects is also a cause for
concern. Those gaps could be addressed by introducing specific provisions in the
Directive” (Ref. 74). Finally, the report touches on problems with harmonising
frameworks for public participation, EIAs with transboundary scopes and the
coordination of the EIA directive with other EU directives and policies (including
the SEA directive).
With respect to my own findings, these two reports provide further support of
the view that the screening procedure is among the main sources for divergences
in EIA practice. It appears fair to assume that divergences between Germany,
Austria and the UK will be negligible compared to countries with much lower
environmental standards; it is these countries that are more likely to raise the
interest of the commission. In the following paragraphs, I will now discuss my
own findings from chapters 3 and 4.
5.2. On the main findings
With respect to the implementation of the EIA into a general clearance
procedure (sub-chapter 3.1), Austria seems to be the country that has adopted
the most efficient and potentially strictest legislation. The centralised and
binding nature of the Austrian EIA proceeding makes the relevant legislation
more comprehensible compared to both Germany and the United Kingdom. This might
be at least partly accommodated by the relatively small size of Austria; however,
it is worth noting that the proceeding itself is administered by the relevant
federal state similar to the situation in Germany or the UK. There, however, the
arborisation of legislation due to traditional (Germany) or newly discovered
(United Kingdom) federalism has led to complex ramifications that constrain the
power of any EIA. I therefore conclude that the Austrian implementation of the
EIA as the core and dominant aspect of a clearance proceeding increases the
significance of the EIA in general.
Sub-chapter 3.2 dealt with four different, more detailed aspects of EIA
legislation in national laws: (1) The objective and the definition of “project”;
(2) screening; (3) the structure of the EIS; and (4) criteria for a clearance.
The findings of this sub-chapter can be discussed individually, even though they
seem to draw a rather clear and coherent picture.
Whilst the definition for “project” seems to follow approximately the same lines
in all three countries examined, Austria dedicates an unusual degree of
attention to the objective of the EIA law. This indicates a very thorough
approach of the Austrian legislative bodies, but does not necessarily lead to
any definite differences compared to Germany or the UK. There, similar
objectives could be derived from other sections of the law. In analogy, there is
no fundamental definition for “project” in the EIA regulations of England and
Wales; nonetheless, very similar developments will qualify as “projects” in the
UK as in Austria (as shown in the section on screening). This “indirect”
approach to circumvent definitions could be explained with the importance of
case law and rulings in the British legal traditions.
As I found screening to the be among the most crucial issues with respect to
discretion problems, and because I have evaluated both legislation and practice
of screening, I will discuss this area separately further down. Looking at
structure and scope of the EIA, the coherence of the national laws with the EIA
directive is remarkable. Even more interesting, however, are the tiny
divergences in the German and Austrian law. Germany explicitly allows
compensatory measures for negative environmental impacts of a project that
cannot be avoided. Even though such measures are common practice in all three
countries, the explicit mentioning of such a policy hints at a more liberal
application of it in the case of Germany. Further investigations into EIA
practice could confirm this. The second German divergence is just a more
detailed explanation of the involvement of the public and can not be a source
for any significant difference to the directive.
The Austrian legislation is different in the outlining of an EIS structure: It
explicitly demands the consideration of construction, operation and
decommissioning (including auditing regime) phase; and is a great deal more
detailed in asking for immission data as well as data on energy demand according
to sources down to the detail of production units (such as machines) and energy
flows. Nothing similarly strict can be found in EIA directive, nor in the German
or British laws. Here we see a major point where the Austrian EIA law goes far
beyond the minimum requirements given by the directive.
Another critical aspect where Austria exceeds both EIA directive and the other
two countries in strictness are the clearance criteria. Here, the EIA directive
makes only very general suggestions – which both Britain and Germany follow.
Austria, where the implementation of the EIA as an independent proceeding makes
the EIA a much more powerful instrument in the first place, defines clearance
criteria in great detail: It obliges authorities to apply best available
technology (BAT) requirements to emissions, immissions and environmental
standards. The way waste is to be managed is also clearly defined. Other
sections deal with formalities, but these few criteria mean that the permission
or non-permission of a project is linked to very clearly outlined conditions –
which is not the case in other member states. Even Germany and the United
Kingdom, both of which have a long-standing tradition in environmental
protection, are a great deal less demanding in this respect.
Looking at practical aspects of EIAs, I quickly found that there are very few
studies on this matter; the ones I found were mostly based on poor data or
applied dubious methods (e.g. descriptions of individual cases). Reliable
numbers are available for the duration of Austrian EIAs, arranged by sector or
type of EIA. Similar studies were almost certainly required by the commission
from Germany and the United Kingdom; however, the 5-year report of the
commission does not reveal any details (Ref. 74) and I did not find any national
reports. The legal basis for deadlines is not very useful, either: As shown in
the case of Austria, the temporal framework outlined in the law is not met by
the actual proceedings and responsible authorities. I therefore have to limit
the discussion of this section to the notion that a detailed study on this
question, maybe using the data reported by the member states to the commission,
could lead to interesting findings.
Looking at the divergences between the legislation of Germany, Austria and the
UK, one can see that in more than one of the evaluated criteria, Austria applies
significantly more rigorous standards than the other two countries. This is
particularly interesting given the fact that neither Germany nor the United
Kingdom are known for lax environmental legislation. In this context, however, I
have to emphasise that the criteria evaluated in detail were not chosen randomly.
I selected criteria that are known to be very strict in Austria and where I
thought that divergences were most likely to occur. One shall be careful to draw
general conclusions about countries being stricter than others based on the data
I have presented in this thesis. There are also a few instances where Austria is
less demanding than Germany or England and Wales. For example, the assessments
of limitations and caveats is explicitly requested in a “concise” statement; a
similar phrase could not be found in the other to national legislations.
5.3. On screening legislation & practice
The comparison of screening legislation is probably the most sensitive area
of EIAs with the greatest potential for harmonisation. Interestingly,
“additional” (not required by EIA directive) projects that are subject to an EIA
that were found in national legislation of Germany, Austria or the UK do not
overlap very much; yet, all of them might be covered by the national laws of all
three countries without being explicitly mentioned. For example, Austria’s
explicit mentioning of particle accelerators does not appear in the German or
the British legislation. Nevertheless, as very large developments, particle
accelerators might still fall under German or British criteria for a compulsory
EIA. Other projects, such as seaside exploration projects, simply do not apply
to countries like Austria for geographic reasons. Cumulative effects of projects
are taken into account in all three evaluated member states; the recent report
of the commission (Ref. 74) must have found the presented deficiencies in this
area in other countries. I did not find drastic differences in the screening
approach or the defined projects in the three countries.
Noteworthy differences in details include Austria’s requirement to assess
projects that involve genetically modified organisms (GMOs). This could reflect
the country’s historically caused aversion against genetic engineering.
Interestingly, the (IMP)3 report’s interviews with EIA stakeholders demonstrated
that including “installations working with GMOs” was – alongside with military
practice grounds and golf courses – among the most commonly given projects for
inclusion in the EIA directive (Ref. 4).
I doubt that the survey done for the (IMP)3 report can be considered to provide
scientific evidence. However, several key-problems in screening practice were
demonstrated and highlighted: “(…) ambiguous screening procedures, found in a
number of cases (lack of transparency in screening decisions, lack of robust
selection criteria); interpretational problems with certain terms and project
type descriptions; demand for adequate reference to the actual impacts on the
environment in setting thresholds [sic] values; and problems in dealing with
cumulative effects” (Ref. 4).
Interestingly, this list was not put together for issues with screening – but as
a summary of the main weaknesses of EIA law in general. The authors recommend a
harmonisation of screening criteria and an extension of the annexes and thus the
list of projects that are subject to EIAs. In fact, this is already a practice
that can be seen ever since 1985 and which is likely to continue.
The study also includes recommendations regarding the application of the annexes
and identifies problems in “…a lack of accurate interpretation of screening
criteria; the need for a closer linkage of thresholds/criteria with the actual
impact; need for more guidance as well as more research regarding EIA (screening)
practice” (Ref. 4). Whilst the first point can be dismissed as somewhat fluffy,
I regard the other two findings to be of key-significance.
Indeed, pollution thresholds often fail to consider environmental conditions
sufficiently; emissions within a certain cap can have very different impacts on
different ecosystems and it is questionable if decisions on this matter should
be left to case-by-case assessments and the judgement of local experts alone (see
sub-chapter 5.4). Furthermore, looking at the poor data that I found on EIA
practice, I fully support the third point quoted above in its calling for
further research on EIA practice.
5.4. Harmony versus Autonomy & further studies
When EIA stakeholders were asked whether they would change the present EIA
directive with respect to the screening framework, the majority of them was
hesitant in making definite statements on any harmonisation (Ref. 4). The
authors of the (IMP)3 report concluded that divergences of EIA regimes were “the
result of different contextual factors – administrative (political), social and
historical – and thus is not necessarily a problem in itself”. This, in fact, is
a neat summary of the main reasons brought up against increased harmonisation of
EIA legislations which are brought up in a rather repetitive manner. It lacks
only the ultimate argument against central criteria: The need to consider local
conditions individually and in a case-by-case manner.
For example, a factory that uses high amounts of water that it pollutes might
fall under the same thresholds in Austria and Andalusia (Spain). However, the
environmental impact it creates might be negligible in Austria, where water is
abundant, but devastating in Andalusia where falling aquifers have been a
problem for many years. Harmonised thresholds for water abstraction or pollutant
load of the affluent would be useless, a case-by-case examination under
consideration of local conditions inevitable. This is exactly what is meant by
linking screening criteria to impacts rather than project features – quite
likely a trend in EIA legislation that will gain momentum in the years to come.
This may also take pressure off the problems that come with high discretion in
the name of subsidiarity and local conditions: As I have demonstrated in this
thesis, there are several aspects in EIAs in which Austria has considerably
stricter criteria than other member states. It would be a highly interesting
question to what extent that imposes a possible comparative disadvantage on
Austria and its standing as a business location. Finding evidence for companies
actually choosing other locations for that reason and quantifying the cost for
the Austrian or any other economy would be very interesting, but
methodologically challenging, if not impossible.
An easier approach for further research in this direction would be as assessment
of case law: Companies might take legal action if screening remains linked to
project features. If these screening criteria are only nationally or regionally
applicable and insufficiently linked to environmental pressures, a distortion of
a market can be the result. Companies that have to operate under strict regimes
might claim their rights regarding equal conditions in a single European market.
Case law of the European Court of Justice could provide interesting insights to
future developments in EIA law. Alternatively, stake holder interviews similar
to those in the (IMP)3 report could be made for comparisons of EIA practice in
individual member states.
Especially differences between member states from Western Europe, Mediterranean
Europe and Eastern Europe would be interesting: These three regions are
historically, economically and geographically/climatically very distinct. I
think it would be reasonable to expect greater divergences between countries
from these regions than those I found in this thesis. One possible starting
point for such an assessment would be the reports of individual member states
that go to the commission. These should include data on implementation,
legislation, as well as practice of EIAs.
Ideally, a thorough study would include a comparison of EISs: The number and
scope of individual opinions; the structure of the EIS; the costs involved and
the criteria that experts have to fulfil for being assigned to them. In Austria,
there are public databases available that contain information on all completed
EIAs (Ref. 75); similar databases could exist in other member states and serve
as useful sources for data. EIAs are complex endeavours, like trees with many
branches: For this thesis, I could follow only a few. Checking on a few more
would be quite interesting, particularly if that would build on my work and
extend the focus beyond the law and further to EIA practice.
5.5. Future directions of EIAs
In its response to the commission’s 5-year report on the implementation of
the EIA directive last year, the Committee of the Regions stated only a few
weeks prior to the completion of this thesis: “(…) in some fields the EIA
Directive is in need of improvement, in particular with regard to screening,
public participation, quality of data, EIA transboundary procedures and
coordination between the EIA and other Directives and policies (…)” (Ref. 76).
This statement is interesting, not only because it was made so recently, but
also because it is based on a very thorough assessment of data on EIA
legislation and practice. Furthermore, it was made by a body with relatively
limited interests in the subject and thus it provided me with an unbiased view
on the matter.
The report is spot-on when it highlights the main areas for improvement and
development with respect to SEA Directive and EIA Directive: (1) Neither
directive has succeeded in establishing obligatory environmental standards; (2)
both directives still have gaps in issues regarding public participation and
transparency; (3) some passages of the two directives overlap; (4) there is a
need to link both directives with issues on biodiversity and climate change; (5)
the screening mechanism in the EIA directive needs to be simplified, annex III
requires new thresholds and criteria, including cumulative aspects; (6)
mandatory scoping and a formalisation of the accreditation of consultants have
to be established; (7) instruments for auditing and documentation of EIAs have
to be developed.
This criticism is very concise and well-grounded; I therefore consider most of
the points mentioned to be key-areas for future developments and reforms of the
current EIA legislation and practice. Based on my own work, I would include a
few more points: (1) I expect a further branching out of the aspects that are
covered by formalised assessments; a separation of health and social aspects in
the form of a mandatory “health impact assessment” could occur as suggested by
the (IMP)3 report (Ref. 28); (2) standardised risk assessment procedures and
qualification standards for consultants could be set up; (3) the EIA procedure
as a whole could be simplified through better implementation into SEAs (the
development of SEA-approved plans and programs is a continuing effort in many
member states and should ultimately lead to a framework within which simplified
EIAs can take place); (4) screening will remain a key-issue and source for
discretion and thus controversy; it is likely that screening criteria will
continue to be developed and will probably be increasingly linked to
environmental standards. Similar developments can be seen in other environmental
directives, such as the water framework directive: There, the ecological state
of a water body has replaced emission thresholds or load caps as defined targets.
The EIA directive is now in place for a quarter of a century. These 25 years saw
a steady increase in the number and sophistication of screening criteria; the
implementation of public participation; and the extension of the scope of EIAs
to transboundary effects. Most importantly, however, the EIA directive provided
a standardised framework that spread to the new member states and candidate
countries, thereby pushing the frontier of mandatory EIAs into a region with
previously very underdeveloped environmental awareness. Based on the work
presented in this thesis, I expect the next 25 years to be easily as dynamic as
return to "CV
(English)" or "Lebenslauf (Deutsch)"