A Curious Case of Port Concessions

The obligations of a concessionaire may be made less onerous in order to enable it to remain in operation. Any adjustment of the obligations takes into account the possible legal defences of the concessionaire.


The Market Economy Investor Principle [MEIP] is a powerful concept. Its many variations attest to its versatility [e.g. market economy investor, operator, vendor, creditor]. It asks a simple but fundamental question. Would a rational economic agent in a similar situation behave in the same way as the public authority that claims to be acting just as another market operator? The question is simple but the answer is never easy and the process by which the answer is derived is in most cases complex and meandering. Normally, the answer is whittled down to a range of numbers because the metric is profit, price, interest, revenue or the amount of recovered loan, depending on the variation of the MEIP that is used.

But how do you apply this principle to a change in contractual clauses? The Commission had recently occasion to examine changes to contractual obligations in decision 2019/1246 concerning two concessionaires in the Port of Antwerp.[1]

In December 2012, the Commission received a complaint from Katoen Natie [KN] alleging that Belgium had granted State aid to two terminal concessionaires, PSA Antwerp [PSA] and Antwerp Gateway [AG], both of which operated in a new terminal of the Port of Antwerp.

The concession contracts had been awarded in 2004 by the Antwerp Port Authority [GHA] after a public tender. The contracts were for 42 years for the operation of container transshipment services in the Deurganckdok [DGD] terminal of the port, which had recently been constructed.

The contracts imposed minimum tonnage requirements [MTRs] whose purpose was “to prevent concessionaires from keeping idle the areas attributed to them.” If they could not fulfil the MTRs, the concessionaires were obliged to pay penalties or compensation to the GHA.

The Commission decision examined two measures. Measure 1 applied to both PSA and AG. It concerned reduction of the compensation payments that had to be made by PSA and AG for failing to fulfil their MTRs. GHA decided to reduce the compensation payments in 2013 with a retroactive effect from 2009 under the so-called “75/125 rule”.

As explained in the Commission decision, “(22) according to that rule, the MTRs were adjusted by adding 125% of the yearly Hamburg-Le Havre range growth rate to the MTR of the previous year. When there was a traffic decline, 75% of the yearly Hamburg-Le Havre range decline rate was deducted from the MTRs of the previous year. The rule thus implies that PSA and AG have to perform 25% better than their peers in the Hamburg-Le Havre range. […[ The ‘outperformance factor’ of 25% is based on the historic performance of the Port of Antwerp, which, in terms of yearly traffic growth rates, over a longer timespan average was 25% better than the average of the ports in Hamburg-Le Havre range. The implementation of Measure 1 led with regard to 2009 alone to MTRs reductions of – 41,7% (PSA) and – 29,7% (AG), when compared to the 2009 original MTRs.”

Measure 2 maintained the reduced compensations [in line with the 75/125 rule] only for the benefit of AG.

The two measures were adopted when the financial crisis of 2008 broke out. PSA and AG were unavoidably affected by the downturn in international and European trade. In those circumstances, if GHA tried to impose the full compensation payments allowed by the contracts, it could drive PSA and AG into bankruptcy.

In fact what happened was that the two companies initiated legal proceedings against GHA in local courts, claiming that even the reduced compensations were abusive because their underperformance was not their fault, as it was attributed to the economic crisis.

In the meantime, GHA gradually reinstated the original MTRs.

Assessment of the existence of State aid

There was only one question here. Did the two measures confer an advantage? Although there was hardly any doubt that all the other criteria of Article 107(1) TFEU were applicable, the Commission did not examine them because it had already concluded that the two companies obtained no advantage. Consequently, there was no need to examine the other criteria.

To establish the presence or absence of advantage, the Commission used the benchmark of the private operator. “(144) In order to determine, for the purpose of Article 107(1) TFEU, whether the reduction of compensations by the GHA granted an economic advantage to PSA and AG, the Commission must establish whether the GHA behaved in a way comparable to that of a private operator in a similar situation. According to the market economy operator principle, a measure carried out by the State does not entail aid where, in similar circumstances, a private investor of a comparable size to that of the bodies concerned in the public sector, operating in normal market conditions in a market economy, could have been prompted to provide to the beneficiary the same measures in question.”

Measure 1

“(146) The Commission notes that, as spelt out in the opening decision, the impact of the Crisis should be taken into account when assessing the market conformity of Measure 1. The Commission further notes that at the time the Crisis started, notably in the years 2008 and 2009, only PSA and AG were subject to progressive MTRs, which increased year to year, while all other concessionaires in the Port of Antwerp, including the Complainant, were subject to fixed MTRs (see recital 29). Therefore, in the context of the Crisis, the situation of PSA and AG was specific and distinct from other operators active in the Port of Antwerp.”

“(147) Moreover, the Commission acknowledges that the unprecedented impact of the Crisis forced undertakings in all sectors of the economy to readjust their behaviour.”

“(148) Furthermore, in addition to the context of the Crisis, the Commission notes that maintaining the cooperation with PSA and AG was important for the long-term interests of the port.”

“(149) The Commission also recalls that KN, the Complainant (see recital 21), and other operators in the Port of Antwerp (see recital 131), were granted MTRs adjustments in their favour in the context of the Crisis.”

“(150) The Commission therefore considers that, as a matter of principle, it is conceivable that a market economy operator in the position of the GHA would also have readjusted the contractually applicable MTRs in view of the Crisis.”

With regard to the specific facts of the present case, “(152) the Commission considered in the opening decision that a rational private market operator would have aimed at maximising its profits (or minimising its losses) by launching the recovery of the sums due as soon as possible. In the case at hand, there was a significant time gap between the start of the Crisis and the 2013 GHA Decision, which retroactively dealt with the issue of unpaid compensations by reducing them significantly.”

Timing of adjustment

“(154) In view of the above, the Commission raised doubts as to whether a rational private market operator would have waited with a decision concerning the reduction of the due compensations for four years, that is from 2009 until 2013.”

“(155) Addressing these doubts, Belgium submitted correspondence between the GHA and PSA respectively AG, showing that as of December 2009, discussions took place with regard to the lowering of the original MTRs. […] As of then, the GHA repeatedly reminded PSA and AG of the unachieved MTRs and asked for their position.”

“(156) The Commission concludes that in the context of the Crisis and in view of the extensive correspondence exchanged between the GHA on the one hand and PSA and AG on the other hand, one could not accuse the GHA of inaction as regards the non-payment of the penalties. Furthermore, the Commission notes that the situation of PSA and AG seemed particularly delicate because their MTRs were progressive (whereas those of other concessionaires were linear) and the DGD was still in the start-up phase (so its operations could have been affected differently than operations in other terminals). Therefore, the fact that the GHA only took a final decision in 2013 does not constitute in itself an indication that the MEO test is not complied with, as a market economy operator might have taken such a decision at a similar point in time.”

Do you know we also publish a journal on State aid?

EStAL banner

The European State Aid Law Quarterly is available online and in print, and our subscribers benefit from a reduced price for our events.


Size of the adjustment

“(157) As to the reduction of the compensation itself, the Commission expressed doubts as to whether a rational private market operator would have granted PSA and AG a reduction of the size granted by the GHA.”

“(159) The Commission recalls that it is conceivable that a market economy operator in the position of the GHA would have reduced the MTRs in the context of a crisis and in view of maintaining long-term cooperation, as indicated in recital 150 of the present decision. However, this does not lead to the automatic conclusion that any reduction of MTRs is market-conform.”

“(161) From the point of view of the GHA several factors are a priori at hand for the sake of adjusting MTRs in the Crisis context. The Commission notes that the GHA applied a combination of two factors. One is the traffic evolution in the Hamburg-Le Havre range and the other one is the downfall in traffic in the DGD.”

“(162) The first factor is defined as the sum of container traffic volumes of all ports in that range. It lies in the nature of a range comprising several ports that some ports perform better, others worse than the range. Each individual dock in each port also typically performs differently than the port average. As a consequence, comparing the performance of an individual dock to the performance of the whole Hamburg-Le Havre range, which is based on the compilation of a relatively big data set, may often lead to inappropriate conclusions. Furthermore, each dock may be affected to a different extent by the Crisis. This speaks in favour of comparing each individual dock against data stemming from a comparable data set.”

“(163) However, the limit of narrowing the data set should be reached at the point where the latter becomes circular … Circularity would result in eliminating the incentive for the concession-holder to achieve certain MTRs because the MTRs would — owing to the circular effect — be retroactively aligned to the Concession-holder’s actual performance.”

“(164) However, the Commission notes that the adjustment applied by the GHA is not circular since it is based on a combination of two factors. One is the Hamburg-Le Havre range and the other is the downfall in traffic in the DGD. The Commission notes that using the Hamburg-Le Havre range as a component of the MTRs adjustment matrix is appropriate as it forms the most likely widest possible data set at hand with regard to operators in the port of Antwerp. … Moreover, the Hamburg-Le Havre range performance is in any event adjusted to the disadvantage of PSA and AG under the 75/125-rule (see recital 21). In fact, the rule implies that PSA and AG must perform 25% better than the Hamburg-Le Havre range.”

“(165) The second factor that formed part of the MTRs adjustment formula was the actual traffic of PSA and AG in the DGD in the year 2008. In that year a 6,3% year-to-year growth of container traffic was observed in the Port of Antwerp. By contrast, from 2008 to 2009, container traffic fell by an unprecedented – 16,3 %. Under the at that time only year-on-year increasing MTRs applicable in the Port of Antwerp, PSA and AG were each held to meet an MTRs in 2009 significantly higher than in 2008, which, when overall container traffic sharply decreased, turned out to be impossible. The Commission holds that it was appropriate to use the realised traffic of each concession-holder in the DGD as the other component of the matrix because otherwise the use of the realised overall port traffic in Antwerp (being the next bigger available example of a narrow range) could have led to inaccurate results. This is so, because the situation in the DGD was very specific since the DGD was still in a start-up phase when the Crisis began.”

“(166) The Commission further notes that the adjustment applied by the GHA resulted in MTRs that in the period 2009-2012 exceeded the tonnages realised by PSA and AG (with the one exception of AG in 2011, when AG performed slightly better than the readjusted MTRs required). Hence the GHA did not waive the penalties due altogether. Also, it is demonstrated that the MTRs adjustment under Measure 1 was not circular, contrary to the Complainant’s allegation. If Measure 1 had been circular, PSA and AG would not have been liable to pay any compensation in the period from 2009 to 2012.”

“(167) Measure 1 remains therefore in line with the ultimate aim of charging penalties for unachieved MTRs, which is not to transform the penalties into a genuine source of income, a return on investment or a payment for services/goods provided by the GHA, but rather to incentivise the concessionaires to perform as well as possible under the given circumstances.”

“(168) The Commission therefore concludes that the methodology used by the GHA for adjusting the MTRs appears to be reasonable.”

Threats to leave the port and risk of litigation

“(173) In the opening decision the Commission also noted that PSA and AG could not credibly threaten to leave the Port of Antwerp in case the MTRs were not reduced or the compensation was not waived (recital 75 of the opening decision). The container terminal operators could not unilaterally terminate their relationship with the GHA, while the GHA could completely or partly revoke the concession in case of manifest underuse on the terminal.”

“(174) Moreover, the opening decision further expressed doubts as to whether a serious risk of litigation from PSA and AG existed (recital 78 of the opening decision).”

“(176) Having analysed Belgium’s submissions following the opening decision and in particular the correspondence between the GHA on the one hand and PSA and AG on the other hand, the Commission holds that there was indeed a serious risk of PSA and AG challenging the enforcement of penalties by the GHA for non-achievement of the original MTRs in litigation or arbitration, as well as a credible risk that such enforcement of penalties by the GHA would be refused by the arbitration panel or by the court under Belgian civil law due to the abuse of law doctrine or due to a penalty clause being qualified as punitive.”

“(177) Based on the contemporary evidence available, the Commission takes the view that Belgian civil law indeed provides the basis to challenge the enforcement of penalties in full for non-achievement of contractual MTRs if such enforcement amounted to abuse of law or the penalty clause were to be qualified as punitive. That is confirmed by the fact that both AG and PSA used arguments related to abuse of law and punitive penalty clause when communicating with the GHA prior to the 2013 GHA Decision, including legal memoranda by their respective legal advisors. While it is not for the Commission to prejudge how effective those claims could be, they proved to be admissible and sufficiently credible, which shows that the risk of litigation could have been perceived by the GHA as real.”

“(179) Furthermore, the Commission notes that the credibility of arguments used by AG in the arbitration procedure against the GHA is confirmed by the arbitration decision in that case. The arbitral tribunal took the view that the enforcement of penalties for non-achievement of the adjusted MTRs under the 2013 GHA Decision did not to amount to abuse of law under Belgian Civil Law. In doing so, the arbitral tribunal took account of the fact that the GHA adjusted the MTRs in its 2013 GHA Decision after taking into account the Crisis, the situation in the Hamburg-Le Havre range and the situation of Antwerp in a historic perspective, and that such behaviour was in conformity to what could be expected from the reasonable and prudent port authority.”

“(181) As regards the risk of litigation, and in particular the risk of unsuccessful litigation for the GHA, the Commission concludes that the proceedings actually initiated by PSA and AG show that this risk was real, and that a prudent market operator would have considered such risk when taking business decisions.”

The Commission concluded in relation to Measure 1 that “(182) as a matter of principle, it is conceivable that a market economy operator in the position of the GHA would also have readjusted the contractually applicable MTRs in view of the Crisis. With respect to the reductions granted by the GHA under Measure 1, the Commission found that (i) the time elapsed between the start of the Crisis and the moment when the GHA took its decision does not constitute an indication that the MEO test is not complied with; (ii) the methodology used by the GHA to determine the size of the adjustment of MTRs appears to be reasonable; (iii) the decision by the GHA did not provide an incentive for PSA and AG to relocate traffic to other ports in which compliance with the MTRs was strictly enforced; (iv) elements such as land-use planning, mobility and long-term employment seemed to have played only secondary role; and (v) that if the MTRs were not adjusted, there was a real risk of unsuccessful litigation that a prudent market economy operator would have taken into account.”

“(183) Taking all these elements into account, the Commission finds, on balance, that Measure 1 must be considered to comply with the MEO test.” “(184) As a consequence, the Commission concludes that Measure 1 constitutes no economic advantage, neither in favour of PSA nor of AG.”

Measure 2

“(185) The assessment whether the GHA, when adopting Measure 2, acted as an MEO, and whether, as a consequence, Measure 2 did not amount to an advantage within the meaning of Article 107(1) TFEU in favour of AG, requires establishing whether maintaining the 75/125 rule with regard to AG resulted in an advantage that AG would not have obtained in case the original MTRs had been reinstated.”

“(186) This question has to be assessed in the light of the fact that in 2014 ca. 30% of AG’s concession area was transferred to PSA. When the GHA reinstated the MTRs for AG in 2015, it deducted an amount corresponding to the 30% concession area loss from the original MTRs to arrive at MTRs of 1 247 630 TEU.”

“(187) The Commission notes that AG was already performing at a level beyond the reinstated MTRs in 2014. In 2015, AG was again performing beyond that level. The question whether a prudent market economy investor would have reinstated the MTRs for AG already in 2014 or only in 2015 (as the GHA did) is therefore not relevant for establishing whether the GHA granted any advantage in favour of AG. No penalties would have been in any event due by AG to the GHA for 2014 and 2015. Under a re-instatement of the original MTRs (duly adjusted by the loss of ca. 30% of the concession area) as well as under the continued application of the 75/125-rule (in line with Measure 2) AG would not have been liable to pay the GHA any compensation.”

“(189) Therefore, the Commission concludes that Measure 2 did not result in an economic advantage in favour of AG.”

A few concluding thoughts

This is an instructive and curious case.

It is instructive because it shows how the MEOP can be applied to changes in contractual obligations and that the options of a public authority that acts as a market operator are circumscribed by what is legally possible and the options which are afforded to its debtors by law.

It is also curious for at least six reasons.

First, like the “curious incident of the dog in the night-time” or the better-known dog in Sherlock Holme’s adventure that didn’t bark, the Commission mentioned the uniqueness of the PSA and AG [paragraphs 146 & 156], which were the only two port concessionaires with progressive MTRs, but then made nothing of it. Because of their MTRs were more burdensome by 25% in comparison to other MTRs, the implicit assumption was that the crisis hit them more severely. But were they also in a position to absorb the impact of the crisis more effectively than other companies in the port of Antwerp? Why was “the situation of PSA and AG […] specific and distinct from other operators active in the Port of Antwerp”? Was this not a factor that should have been taken into account? Did their position in the port confer a certain privilege or advantage [e.g. better facilities or more efficient processes in the new terminal]. The Commission mentioned several times that the new terminal was in its start-up phase, but again it did not draw any explicit conclusions. And, if the terminal was in its start-up phase, were the companies in their start-up phase too or were they well established with long experience in managing port terminals under different economic conditions? [The fact that there was a risk they could move their business to other ports suggests that they were well established companies.]

Second, there was no comparison between the adjustments in favour of PSA and AG and those for the benefit of other companies. Were the adjustments for PSA and AG similar to other adjustments? Were they more or less favourable?

Third, the Commission considered the reliability of the Hamburg-Le Havre benchmark and, correctly, observed that a good benchmark compares like with like and that if the benchmark is too narrow you end up with comparing a company to itself [paragraphs 162-163]. In the words of the Commission, the benchmark becomes “circular”. The art and science in constructing a benchmark is to include all the relevant factors and to exclude the irrelevant. The Commission implied that the Hamburg-Le Havre benchmark was too broad, but that an index based on traffic at the DGD terminal would be circular. However, it went on in paragraphs 164-165 to say, without indicating the weight that was assigned to the Hamburg-Le Havre benchmark, that the adjustment made by GHA was not circular because GHA used two factors: the Hamburg-Le Havre traffic and the traffic in the new terminal of the port. It is not obvious how the combination of the two factors removed the circularity. In paragraph 166, the Commission made a smart statement. If the benchmark were circular, PSA and AG would have paid not penalties. In reality, however, the use of the second factor – the development of traffic in their own terminal – meant that their underperformance was minimised. Had it not been used, the penalties would have been larger. Given that the size of the penalties is determined by their performance in relation to a benchmark, the question as to the correct benchmark is still open.

[Unless one can rely on a broader principle or criterion to discard irrelevant factors or influences, benchmarking of performance is a hopeless exercise in practice. A more credible approach is econometric analysis by which the various factors that can affect performance, such as size, etc, can be quantified and a predicted value can be derived. Then the actual performance can be compared to the predicted performance.]

Fourth, if “maintaining the cooperation with PSA and AG was important for the long-term interests of the port”, what would have been the financial impact on the port had PSA and AG gone bankrupt or ceased operations? Could they not be replaced by other operators? As is noted in the decision, GHA could unilaterally terminate the concession agreements, while PSA and AG could not. This implies that GHA was in a strong negotiating position. On the other hand, if PSA and AG were important to the port, then they were in a strong negotiating position. These opposing but important elements seem to have been ignored.

Fifth, GHA faced a risk of litigation by PSA and AG that its actions would be considered as abusive by Belgian courts. GHA did not have to impose abusive penalties. It is conceivable that even if the penalties were found to be abusive, they could be subsequently reduced. The risk of litigation did not mean that no penalties could be imposed. It only meant, I think, that there was an upper limit to the compensation that could be demanded by GHA. The Commission dealt extensively with the risk of litigation in its decision, but did not make anything of the fact that the two companies had actually lost in court.

Lastly, the MEIP requires a public authority to put itself in the position of a market operator. I think that if a market operator had to determine the right amount of penalties, it would start by calculating the maximum amount allowed by the concession agreements and then reduce it by taking into account what PSA and AG could afford to pay without going bankrupt, without moving business to another port and without being able to claim that the penalties were abusive. This requires some quantification of the possible options. Perhaps such a quantification did take place but it is not apparent in the decision, nor is it indicated whether indeed that was the methodology used by the GHA.


[1] The full text of the Commission decision can be accessed at:


[Photo by Steve Doig on Unsplash]



Phedon Nicolaides

Dr. Nicolaides was educated in the United States, the Netherlands and the United Kingdom. He has a PhD in Economics and a PhD in Law. He is professor at the University of Maastricht and the University of Nicosia. He has published extensively on European integration, competition policy and State aid. He is also on the editorial boards of several journals. Dr. Nicolaides has organised seminars and workshops in many different Member States, and has acted as consultant to several public authorities.

Hinterlasse eine Antwort

Zusammenhängende Posts

18. Aug 2020
State Aid Uncovered von Phedon Nicolaides

MEIP-Compliant Bank Recapitalisation: CEC Bank

An investor who is already a shareholder would take into account not only the return on new investment but also the impact on the overall profitability of the company in which capital is injected. Update on Temporary Framework: Number of approved and published covid-19 measures, as of 14 August 2020: 252* Legal basis: Article 107(2)(b): 27; Article 107(3)(b): 211; Article […]
26. Nov 2019
State Aid Uncovered von Phedon Nicolaides

How to Assess a Shareholder Loan

A shareholder loan cannot be simply compared to a bank loan because the shareholder also benefits from improvements in the future profitability of the borrower. Introduction When a public authority grants a loan to a company, the methodology in the 2008 Commission communication on reference and discount rates can be used to determine whether the loan contains State aid. However, […]
09. Okt 2018
State Aid Uncovered von Phedon Nicolaides

Ex ante Assessment of Future Profitability is Absolutely Necessary

A private investor assesses the prospects of future profitability before it invests. The burden of proof lies with the Member State that claims it has acted as a private investor.   Introduction The market economy investor principle is based on a simple premise: before you commit your money you need to check how much you are likely to get back. […]
13. Feb 2018
State Aid Uncovered von Phedon Nicolaides

Investment in Airport Infrastructure

Public funding of infrastructure used for non-economic purposes does not constitute State aid. Public investment in infrastructure used for economic purposes does not constitute State aid if the investment can generate a return that is acceptable to a private investor.   Introduction   On 25 January 2018, the General Court ruled in case T-818/14, Brussels South Charleroi Airport v Commission.[1] Brussels […]
02. Jan 2018
State Aid Uncovered von Phedon Nicolaides

PART II: Concessions and State Aid: Does the State Act as a Regulator or Market Operator?

When Member States act as regulators they need not maximise revenue from the granting of concessions rights. When Member States act as regulators they must grant concession rights on the basis of procedures which are competitive, transparent, non-discriminatory and unconditional.   Part II: Public procurement principles[1] Principles for selection of concession holders The Commission examined also, in addition to the […]
27. Dez 2017
State Aid Uncovered von Phedon Nicolaides

PART I: Concessions and State Aid: Does the State Act as a Regulator or Market Operator?

When Member States act as regulators they need not maximise revenue from the granting of concessions rights. When Member States act as regulators they must grant concession rights on the basis of procedures which are competitive, transparent, non-discriminatory and unconditional. Introduction   Is it possible for a public authority to grant State aid through a defective procurement procedure? The answer […]
05. Dez 2017
State Aid Uncovered von Phedon Nicolaides

Imputability of an Aid Measure to the State Does not Require a Counterfactual

The fact that the state owns an undertaking is not enough to prove that the decisions of that undertaking can be attributed to the state. However, it is sufficient that the state was involved in the particular decision that transferred state resources for the benefit of another undertaking.   Introduction Several recent articles on this blog have examined the concept […]
12. Sep 2017
State Aid Uncovered von Phedon Nicolaides

Concession Agreements Free of State Aid

Awards of concession contracts should follow a procedure which is open, transparent, non-discriminatory and unconditional. Prices may be regulated to prevent excessive prices and profits. But, the concession fee may be adjusted to enable the concessionaire to earn a sufficient return.   Introduction This article examines the link between State aid and concession agreements. In decision SA.44259, the Commission considered […]
29. Aug 2017
State Aid Uncovered von Phedon Nicolaides

Bank Recapitalisation that Conforms to the Market Economy Investor Principle

A public authority acts like a private investor when it injects capital in a stricken bank if there is a realistic prospect of sufficient return that compensates it for the risk it bears.   Introduction Caixa Geral de Depósitos (CGD) is the largest bank in Portugal and is now fully owned by the State. In June 2012, Portugal notified to […]
20. Jun 2017
State Aid Uncovered von Phedon Nicolaides

The Effectiveness of Clear Guidance: The Case of Broadband Networks

This blog examines how much aid goes to support broadband networks and what lessons can be drawn from the Commission’s decisional practice.   Introduction A principal objective of the State Aid Modernisation was to free Commission resources from the time-consuming task of checking the conformity of routine measures of State aid. Consequently, the General Block Exemption Regulation was extended to […]