External Quality Assurance of the Fehmarn Belt Fixed Link Business Case—Too Little, Too Late, and Too Unreliable

The external quality assurance (EQA) of the Fehmarn Belt fixed link business case commissioned by the Danish Ministry of Transport is critically analyzed regarding the New Construction Budgeting requirements as well as common practical criteria. A time line of analyses, decisions and external quality assurance activities as well as external criticism is established as a basis for analyzing the completeness, timeliness and trustworthiness of the quality assurance. Information obtained per the Public Records Act is included. It is found that the external quality assurance was grossly incomplete with limited scopes and exclusion of major elements of the project. The overall financial performance and financial uncertainty analyses were not subject to external quality assurance at all. The bulk of external quality assurance was carried out only after the decision-to-build was made, preparatory construction activities initiated or major economic commitments made, thus jeopardizing the timeliness. The trustworthiness of the external quality assurance is limited for different reasons: bad timing of activities, lack of independence of consultancies, and controversial findings of opposing analyses. The external quality assurance is found to be too little, too late, and too unreliable. The status of the project as a high-risk business case is not improved by the external quality assurance.


Introduction
The Fehmarn Belt fixed link project, currently put on hold due to lacking German approval, has a long and turbulent history, of which a short account is giv-en (in the Appendix, a detailed timeline is presented). Way back in 1991 in connection with the Danish-Swedish agreement to build the Øresund fixed link the project was put on the official political agenda of the Danish Parliament. Denmark announced that the possibility of a fixed link crossing the Fehmarn Belt should be further investigated as an important element in the direct connection between Stockholm and Paris. Some of the early investigations may be found in FTC (1999), TRM (1999) and Femern A/S (2003). In 2000, the Ministries of Transport of Germany and Denmark launched an inquiry of commercial interest (ECI) to investigate the private sector's interest in implementing a fixed link for road and railway traffic across the Fehmarn Belt, State Memorandum (2000), and the results were published, FDJV (2002). The forecasted income from the project was considered too low and uncertain to support a private sector investment, in part due to the competition from other links, such as ferry services and the Great Belt fixed link. The conclusion from the ECI is that the project can only be realized with substantial public support either in the form of Only ten days later, yet another memorandum was agreed upon, State Memorandum (2007). However, the German government had already decided to pull out, except for building and financing the German hinterland constructions, State Treaty (2008). The treaty followed a binding political agreement between a majority of Danish political parties, Danish Parliament (2008). Still, the environmental approval of the project part on German soil must be given by the local state government of Schleswig-Holstein, or more precisely by the organization: Landesbetrieb Strassenbau und Verkehr Schleswig-Holstein. The agreement, Danish Parliament (2008), only two pages long, was unconditional. It merely stated the political intention to build, finance and operate such a fixed link without mentioning any economic or other conditions to be fulfilled. At the time, some initial financial analyses had been done, Femern A/S (2003), TRM (2004), based on a recent traffic forecast, FTC (2003). Also, external criticism was exercised prior to the political agreement, Jespersen (2007) and Vieregg-Rössler GmbH (2008). A new financial analysis was carried out, Femern A/S (2008), however still using the traffic forecast from 2003.
As from 2008 the Danish political parties behind the project have continued the struggle to implement the project despite the lack of commercial and German political support (except for the hinterland constructions) and severe public criticism from sceptics. Following another binding political agreement on a green transport policy, now also including the Danish People's Party in the Fehmarn Belt agreement, Danish Parliament (2009a), a planning act was passed that allowed for the planning process to continue with increased intensity, Danish Parliament (2009b). At the same time the State Treaty (2008) was ratified. The rivalry between a cable stayed bridge, originally preferred, and a submersed tunnel was ended with a victory for the immersed tunnel, Femern A/S (2010aFemern A/S ( , 2010bFemern A/S ( , 2010c, decided by the politicians behind the fixed link on 1 February 2011, Femern A/S (2011a), and calculated by Femern A/S (2011d). A proposal to locate the tunnel element factory in Rødbyhavn, Femern A/S (2011c) was also agreed on, TRM (2011b). Construction activities were advanced by the approval of the Ministry of Finance, Danish Parliament (2011), and consolidated construction costs of the immersed tunnel were put together, Femern A/S (2011d). To advance the bidding process for the tunnel construction works the Ministry of Finance had to approve of yet another act, Danish Parliament (2013).
At the time, there was a lot of optimism concerning the implementation time schedule of the project, Femern A/S (2012). It was planned that the application for German approval would be submitted by April 2013 and the approval obtained by January 2015. Resolving of potential court cases was not considered. As of today, the approval is expected by 2020 at the earliest, including the resolution of court cases. The building process was expected to take 6.5 years, today 8.5 years are expected. Initially, the fixed link was expected to be inaugurated by 2018, State Treaty (2008), now the expectation is 2028, at the earliest, Femern A/S (2016a).
In Section 2 an outline is given of the concept of New Construction Budgeting initiated by the Danish Ministry of Transport in 2006, and how and to what extent external quality assurance should be applied to the Fehmarn fixed link project. In the following sections the paper deals with quality assurance of elements of the Fehmarn Belt fixed link project. Section 3 covers road traffic forecast quality assurance, which is of critical importance for link economy. Special attention is devoted to the question of assumed road traffic transfer from the Great Belt and discontinuation of the existing privately owned ferry service crossing Fehmarn Belt. Section 4 focuses on quality assurance of the construction costs of hinterland constructions, tunnel construction and allocated reserves and the delayed German approval process, including the contractual consequences. In Section 5 the consolidated financial analysis is considered including financial uncertainties. Finally, the conclusion and references follow. This paper was completed by 18 March 2017. It is a revised version of Schjaer-Jacobsen (2017b). Quite a substantial part of the information and documents used in the paper has been obtained while the author and others were granted access to public records in accordance with the Public Records Act. A few access applications are still pending.

New Construction Budgeting, Quality Assurance and Reserves
After a couple of years with many serious cost overruns on projects run by the 1) External quality assurance carried out by an external and independent consultancy.
2) Experience-based correction supplements which are percentage reserves added on the cost side.
The aim is to improve cost control and prioritizing of projects. The principles should be used in all larger infrastructure projects within the domain of the Ministry of Transport and apply to the appraisals at the two points in the decision process labeled Level 1 and Level 2. At Level 1 it is decided which projects are taken forward and at Level 2 the decision-to-build is made. The standard for experience-based correction supplement is 30% of the base budget, TRM (2010a).
In the main memorandum, TRM (2010a), the principles for implementation are laid out. Terms of reference for the external quality assurance at Level 1 is found in TRM (2010b) and Level 2 in TRM (2010b). We will focus on Level 2 external quality assurance that should be carried out as a basis for the political decision-to-build to be taken by Parliament, usually by passing a construction act. Obviously, it is seen from Table 1

Road Traffic Forecasts
The consulting company COWI was chosen to perform external quality assurance of the road traffic forecasts. Since COWI has a major interest in the realiza- During April and May 2015, the work program, COWI (2015a), for an external quality assurance of the road traffic forecasts was negotiated between The Ministry of Transport and COWI simultaneously with the preparatory work of the construction act. The method adopted for the quality assurance was New Construction Budgeting, TRM (2006aTRM ( , 2010aTRM ( , 2010bTRM ( , 2011a. When the report came out, COWI (2015b), the initiative had not been announced publicly, and the timing was peculiar considering that the subject of traffic forecasts was not considered to be critical in relation to the passing of the construction act, Danish . The early traffic forecasts, FTC (1999FTC ( , 2003, had been criticized by Jespersen (2007) and Vieregg-Rössler GmbH (2008)  The external assurance report, COWI (2015b: p. 9) has a main conclusion in two parts (translated by the author): "The main conclusion is that COWI assesses that the traffic forecast of the main scenario is a realistic estimate of the traffic volume on a Fehmarn Belt fixed link provided the ferry operation Rødby-Puttgarden is discontinued. The forecasted growth over time corresponds to the historical growth in later years. The assumptions are considered reasonable and the forecast models are consistent with the professional practice of traffic forecasts. However, we think that there are elements in the forecast that appear to be difficult to substantiate. Particularly, the expected transfer of passenger car traffic from the Great Belt is difficult to substantiate due to a lack of data concerning the actual traffic pattern crossing the Great Belt fixed link. The forecast is conservative in estimating the creation of new traffic due to the lack of tools for forecasting the potential dynamic effects of the fixed link". In the report COWI does not address several fundamental questions of great importance for the trustworthiness of the traffic forecasts, such as the inability of the forecasting method, FTC (2003) and Intraplan/BVU (2014a, 2014b, applied to handle the complexity created by two different segments of passenger cars (Europe travelers and border shoppers) and two competing transport modes (tunnel and ferry). Likewise, the lack of evaluating the degree of uncertainty in data for the traffic forecasts is ignored. However, the estimated transfer of passenger car traffic from the Great Belt is explicitly addressed. This is particularly important because the estimated volume amounts to approximately 800,000 one-way cars per year, corresponding to approximately 25% of the expected tunnel traffic.
Despite the reservation in the main conclusion and the weaknesses mentioned above the Minister of Transport is quoted as saying on 12 November 2015 (translated by the author): "COWI has found that the traffic forecast is thorough and presents a realistic estimate of the tunnel traffic. This is an important conclusion and then we politicians can concentrate on the remainder of the project", TRM (2015c). The minister's omission was pointed out, Bredsdorff (2015b) and Schjaer-Jacobsen (2015b), but even more interestingly: On the very same day, namely 12 November 2015, as the Minister of Transport approved the traffic forecast quality assurance, it turned out that the Ministry of Transport requested an analysis, TRM (2015d)  The latter concludes: "In our opinion, there are significant uncertainties related to conclusion in the KPMG report that the Rødby-Puttgarden ferry service be loss-making from 2026. We consider it more likely that the Rødby-Puttgarden ferry service will continue to be profitable in 2026". To the author's knowledge the only attempts to base this discussion on a genuine competitive theory were DIW Econ GmbH (2015c) and Aigner (2016). Particularly the latter suggests that the ferry is a much stronger competitor to the fixed link than assumed and that one should not take it for granted that the ferry will exit the market. In fact, it seems more likely that the ferry will make positive profits in equilibrium and stay in the market. This is quite a challenge for the fixed link because the results also suggest that if the ferry competes, the tunnel will not be a profitable business, Aigner (2016). New Construction Budgeting, see Table 2.

Reserves and Risk Distribution of Tunnel Constructions
Otherwise, see Table 2 for the coast-to-coast tunnel construction costs. Even though New Construction Budgeting did not apply, two alternative calculations were made in 2014, one with reserves of only 14% and one with 30%, Femern A/S (2014b). The former was based on estimates from the tunnel construction consortia, the latter on requirements from New Construction Budgeting. Only a few months later the construction costs exclusive of reserves had risen from DKK 40.5 to 49.4 billion due to new incoming bids from the tunnel construction consortia, Femern A/S (2015b), increasing the payback period to 39 years. This was essentially the financial calculation behind the presentation of the construction act L141 to . However, tunnel construction costs were increased by DKK 2 billion and reserves reduced by DKK 1.8 billion, see Table 2. A brief status report was issued at the same time, Femern A/S (2015c). The construction act was passed conditionally, meaning that construction work startup had to await a negotiated reduction of construction costs, a clarification of the amount of EU subsidies that could be obtained, and an acceptable time schedule for the German environmental project approval. It turned out that negotiations with the tunnel construction consortia resulted in lower bids, a prolonged construction period, and a redistribution of risks between the master builder and the construction consortia, Femern A/S (2016b). The question about EU subsidies was clarified by 29 June 2015, resulting in much lower subsidy than expected, Børsen (2015). A time schedule for the German approval process has not yet been presented by the Germans, Femern A/S (2016f).
On 1 October 2015, the Danish Ministry of Transport decided to have a "due  Parliament (2009a), decided to commission an external assessment and quality assurance of the Fehmarn Belt fixed link reserves and risk distribution between the master builder and the main contractors of the tunnel construction works, including a perspective of the delay of the German approval process, Ernst & Young (2016). This initiative was taken after the construction act was conditionally passed by the .
The task was to answer four questions (translated and edited by the author): 1) How are today's risks distributed between the master builder and the con-struction consortia and how were they distributed earlier? 112 risks were identified, 9 of which changed the risk allocation and 12 caused a change of Fehmarn A/S's risk exposure. Only 10% -15% of the reported cost reduction is caused by change of risk distribution while 85% -90% is caused by a reduction of tasks and an increase of the building period.
2) Does the risk distribution correspond with the common praxis of construction contracts for large projects? The actual distribution is comparable with international standards of other large international projects. The risk distribution should be supported by an improved risk management process.
3 The statistical approach applied by Ernst & Young (2016) to arrive at a forecast of the date of the German approval is criticized, Ingeniøren (2016d). Actually, the approval has already been delayed by another 6 months since the forecast was published, TRM (2016c). Consequently, it is highly likely that construction work cannot be started before medio 2020. The delay of the German environmental approval has got the political attention of the Danish Minister of Transport. Bypassing the political parties behind the project as well as the state government of Schleswig -Holstein he addresses the federal government in Berlin directly to find ways of speeding up the approval process, TRM (2017b).

Financial Analysis and Uncertainty
The is also the case elsewhere -that as long as we are under 40 years we are in the green zone, then we are on safe ground regarding the taxpayers avoiding to pay.
When we are between 40 and 50 years we are in the yellow zone, then it begins to be on shaking ground. When we are over 50 years we are in the red zone and we must stay completely out of that one, that is what I mean", quoted from Schjaer-Jacobsen (2016b, 2017a).
Based on alternative calculations, Bredsdorff (2015a), Schjaer-Jacobsen (2015a), Rasmussen (2015), the finances of the Fehmarn Belt business case were criticized prior to the passing of the construction act on 28 April 2015.
Although of immense importance for the financial feasibility of the fixed link project, the external quality assurance efforts carried out as described in this paper did not address such vital items as: • Inability of the traffic forecasts methods to handle a business case subject to competition. • Transfer of road traffic from the Great Belt to the tunnel (quality assurance requested 16 months ago).
• Uncertainty analysis of road traffic volume.
• Tariffs for tunnel road traffic.
• Uncertainty analysis of road traffic income.
• Amount of EU subsidies.
• Economic consequences of needed renegotiations of main tunnel construction contracts.
• Maintenance and administration costs of the fixed link.
• Consolidated financial analysis of the business case.
• Financial risk and uncertainty analysis of the business case.
Genuine uncertainty analyses beyond simple partial sensitivity analyses applied by Femern A/S can be found in Table 3. By comparison with Table 2 it turns out that the financial uncertainties are much larger than the impression left by the official financial reports conveyed to Parliament and the public.
Clearly, the likelihood that the Payback Period will be in the green zone is practically equal to zero.
Consequently, the external quality assurance was rather incomplete leaving large areas of continued doubt concerning the financial performance of the fixed link. years (see Table 2) and the external quality assurance of reserves and risk distribution, Ernst & Young (2016), was published, a lot of things happened, for example: • Tunnel contractors have been appointed, Femern A/S (2016c), and conditional tunnel construction contracts have been signed, Femern A/S (2016d).
Thus, extra costs will be incurred due to stand-by fees until construction start-up and additional costs for keeping Femern A/S operating over an extended period, estimated in total to DKK 0.5 billion per year. • A renewed application for German approval has been completed and submitted, Femern A/S (2016e), replacing the previous one, Femern A/S (2013).
This has incurred extra costs.
• The German approval process has been delayed by another 6 months, TRM (2016c). Consequently, the tunnel construction contracts must be renegotiated because the German approval will probably not be obtained before the contracts expire. The amount of extra costs is unknown.
• Answers to 12,600 German objections to the fixed link projects have been completed and submitted, Femern A/S (2017a). This has been a cumbersome and costly affair, costs being unknown.
• Scandlines announces long term competition with the Fehmarn Belt fixed link by upgrading the Gedser-Rostock ferry connection, Børsen (2017a). This initiative may cut away a substantial part of the forecasted traffic transfer from Gedser-Rostock to the Fehmarn tunnel. ently, all the above-mentioned potential costs are expected to be accommodated in the budget without depleting the reserves and jeopardizing the economy of the project. Particularly notable is the absence of any kind of risk or uncertainty analysis, not even partial sensitivity analyses are included. In the author's opinion, the Fehmarn Belt fixed link business case is still a financial high-risk project, as pointed out by Schjaer-Jacobsen (2016a, 2017a), despite the efforts of the project owners to rationalize the decisions and the economic commitments made by performing external quality assurance.

Conclusion
In summary, the external quality assurance dealt with in this paper includes the following official reports (in chronological order) as well and the contexts in which they were created, published, and received: • KPMG (2012). EQA of railroad constructions.
• KPMG (2016). EQA of competition by ferry services. The paper finds that the quality assurance concerning the Danish land works is carried out in compliance with the principles laid out in New Construction Budgeting concerning completeness, timeliness and trustworthiness. This is not the case concerning the tunnel construction work and the entire business case including Danish land works. Paradoxically, New Construction Budgeting is claimed not to apply but nevertheless it is invoked, however in general not complied with. The quality assurance commissioned and carried out is not complete, since large and important issues are not dealt with and left out of the analyses. The quality assurance is not timely since it is carried out at times when decisions and large economic commitments have already been made.
Largely, the quality assurance is limited in trustworthiness, partly due to bad timing, partly due to lack of independence of consultancies, and partly due to findings of opposing analyses. In other words, the external quality assurance of the Fehmarn Belt fixed link business case is too little, too late, and too unreliable.
Thus, the Fehmarn Belt fixed link project is still a high-risk business case, which contrasts with the impression created by official communications. The political majority of the Danish Parliament has decided to continuously promote the Fehmarn Belt fixed link project despite the large likelihood of a financial project failure. Generally, the external quality assurance efforts have not contributed to verification of the financial viability of the project.

Appendix: Timeline
Fehmarn Belt fixed link analyses, decisions, external quality assurance (EQA), and external criticism. Extracted from References.