1. Simulation and its consequences in European contract law
Not infrequently the parties to a business transaction explicitly conclude an agreement while tacitly consenting that this agreement does not reflect their true intentions and will not be implemented as it stands. This is the phenomenon of simulation in contract law. Many legal systems make a distinction between absolute and relative simulation. A fictitious marriage, which is often concluded in order to procure a residence permit for one partner, provides an example of absolute simulation. The partners do not have the intention to live together as a married couple. In such cases of absolute simulation the parties conclude an (apparent) contract which they do not in fact want. In cases of relative simulation, the overt conclusion of an unwanted contract goes together with the secret conclusion of a second contract which conforms to the true intentions of the parties. The simulated contract serves to cover the other contract (the dissimulated contract) which the parties intend to put into practice. Parties resort to the technique of relative simulation when the contract that they actually want to conclude is legally prohibited. For example, in Roman law a husband would pretend to sell a thing to his wife and both would agree on a price. In fact, the husband would donate the thing to his wife and the price would never be paid. In this way, the couple sought to conceal the fact that they violated the prohibition against donations between married partners which existed in Roman law (and still exists in French law). In modern law, written contracts for the sale of real property often contain a price that is lower than the amount that will in fact be paid by the buyer. Thus, the parties attempt to evade part of the taxes and fees due for the transaction.
In other cases, the (relative) simulation is not used to cover up an illegal agreement, but to conceal a legal agreement which the parties feel might be damaging to their reputation or their business interests. A dealer who is forced to sell off parts of his stock cheaply may enter into a fictitious agreement to sell for a higher price and secretly agree to deliver at a cheaper price.
Simulated contracts and transactions are void. This basic rule was recognized in Roman law. The sources of Roman law also show that the rule is not designed to punish the parties for their fraudulent acts, but to make their true intentions prevail. Only a contract that is an actual expression of the parties’ will can be legally binding. An agreement that is only intended to look like a binding contract but that is not supported by the will of the parties cannot be enforced. Consequently, under Roman law, the dissimulated contract which was really wanted by the parties and for which the simulated contract served as a mere cover, is valid and binding—as long as it is not prohibited by special legal provisions such as the provision against donations between husband and wife.
The basic rule of Roman law according to which the simulated act is void while the dissimulated act may be valid provided the starting point for the development in the ius commune. All continental legal systems developed a doctrine of simulatio which was based on the Roman sources. The English common law alone failed to deal explicitly with the problem of fictitious contracts. In the age of enlightenment, there was a tendency among legal scholars to regard as void both the simulated and the dissimulated contract. This would have engendered a transformation of the rules governing simulatio into provisions punishing the parties for their lack of honesty. However, the traditional view prevailed.
The basic rule that simulated contracts are void but that a dissimulated contract may be valid is not sufficient to resolve all legal issues arising from simulation. The concept of simulation needs to be clarified. Throughout most of its history, the concept of simulation was not distinguished from the circumvention of legal provisions (fraus legis). Only in the 19th century, legal scholars started to treat simulatio and fraus legis as separate phenomena subject to different regimes.
In addition to that, the exact scope of the rule that simulated contracts are void must be determined. Beginning in the 18th century, legal scholars debated whether the contracting parties should be allowed to assert the nullity of the simulated contract against unsuspecting third parties. The French Code civil contains provisions to protect third parties, while the Prussian (Allgemeines Landrecht für die Preußischen Staaten (ALR)) and Austrian codes (Allgemeines Bürgerliches Gesetzbuch (ABGB)) and later the German civil code (Bürgerliches Gesetzbuch (BGB)) left this question to courts and legal scholars.
2. Trends of legal development
All modern codes contain the basic rule that the simulated contract or agreement is necessarily void while the dissimulated one may be valid depending on the circumstances. The proposal of reforming the law in order to deter the parties from entering into fictitious agreements by either refusing to enforce the dissimulated contract or by binding the parties to the simulated contract is sometimes floated in scholarly literature. However, the traditional rule still prevails throughout Europe.
As was already mentioned, English law apparently never developed a general theory of simulation, but rather follows the general rule that contracts have to be construed in accordance with the true intentions of the parties. However, the concept of a ‘sham transaction’ which is employed by common law courts in tax and labour law cases resembles the continental doctrine of simulation. Moreover, the traditional maxim of equity, ‘equity looks at the intent rather than the form’, expresses the idea which forms the basis of the continental approach to simulation.
There seem to be no statutory provisions designed to clarify the distinction between simulation and related legal phenomena such as fraus legis. In many countries the distinction is made by asking whether the parties only intend to create the appearance of a contract which they do not in fact want to implement, or whether they enter into a transaction which has to be valid in order to serve its purpose. According to this distinction, there is no simulation in many cases in which the parties resort to the use of a straw person. For example, in German law (and in some other systems), a pledge can only be validly given if the thing to be pledged is physically controlled by the creditor, while the right of ownership can be transferred without a change of physical control (possession). If a debtor transfers his ownership rights in a machine to a creditor to provide the creditor with a security, the transfer of ownership may not be qualified as a simulated conveyance dissimulating the giving of a pledge. The parties actually want ownership to pass. If ownership does not pass, the purpose of securing the debtor will not be fulfilled. Whether such a transaction is valid or whether it is void because it serves to defeat the provisions on pledges has to be determined on the basis of the rules governing the giving of pledge. The rule that simulated legal acts are void is not applicable.
The sharp distinction between simulation and evasion of the law which is prevalent in modern law can be traced to German (pandectist) legal scholarship in the 19th century (historical school). Prior to that, simulatio and fraus legis had been treated as (almost) identical phenomena for many centuries. The distinction was clarified as the approach to evasion changed. By interpreting laws liberally and purposively, it became possible to qualify many instances of (attempted) evasion of statutory provisions as direct violations of the statute in question. Thus, it was no longer necessary to resort to the doctrine of simulatio in order to regard such attempts at evasion as null and void.
However, the criterion for this distinction was not invented by the German scholars. They only emphasized an idea which was already present in the writings of late medieval commentators of the ius commune. Although the criteria for the delimitation of simulatio and fraus legis is internationally accepted today, difficult cases continue to arise.
It is also universally accepted today that third parties who rely on a simulated contract need some degree of protection. Many countries—like Austria—followed the French model and created statutory provisions for the protection of third parties. Sometimes, these provisions only say that the nullity of a simulated contract cannot be asserted against a third party who relied on the contract in good faith. Other legal systems provide that the third party may at his choice treat the contract as valid or assert its nullity. But the third party is not left without protection even where no statutory provision exists. In many cases, the general rules, eg on the bona fide purchase of property, afford adequate protection. In Germany, the courts have sometimes extended the protection of third parties to cases not covered by any statutory provision. In doing so, the courts continued a line of jurisprudence beginning with cases decided before the BGB came into force and originating in the areas on the western bank of the Rhine where French law was applicable until 1900 (see RG 1 November 1887, RGZ 20, 336). Thus, the French provisions for the protection of third parties indirectly influenced the development of German law even after French law was no longer applicable in Germany.
The greatest difficulties arise in connection with agreements in writing or subject to special provisions regarding their form. In English law, the parol evidence rule prevents the courts from admitting evidence designed to show that the intentions of the parties are not adequately reflected by their written agreement. The parol evidence rule is inapplicable, however, if the written document conceals the true character of the contract (Madell v Thomas & Co  1 QB 230 (CA)). In such a case, the true agreement can be enforced in court even though it cannot be deduced from a written document. Italian law contains a rule similar to the parol evidence rule of the common law. In general, the Italian courts take the view that no party may call witnesses to prove that a written document records a simulated contract and does not reflect the parties’ true intentions. According to the prevailing opinion among the courts, however, this rule is inapplicable if the general character of the contract in question is clear and the alleged simulation only concerns details such as the exact amount of the price to be paid.
This line of jurisprudence is relevant in the case of a written contract for the sale of real property which understates the agreed price in order to facilitate tax evasion. As already mentioned, this case is of particular practical importance in Italy and elsewhere. The position of the Italian courts makes it possible for the seller to call witnesses in order to prove that in fact a higher price was agreed upon. There is a judgment of 2004, however, which departs from this rule and, by declaring the evidence offered in support of the alleged higher price inadmissible, effectively binds the parties to the simulated agreement (Corte Suprema di Cassazione, sezione prima civile, 19 March 2004, n. 5539, Foro Italiano 2005, 510).
In France, there is a special provision in the tax code which dictates the same solution in the special case of a contract for the sale of real property understating the price. In Germany, there is no restriction of the means by which a simulation can be proven. However, in the case of real property sales contracts understating the price, the dissimulated contract for the sale at a higher price is regarded as void by the German courts because this contract is not contained in a notarial deed. Swiss courts decide similarly, while in Austria the sale for the higher price that the parties actually agreed upon is regarded as valid and binding.
It is worth noting that in the case last mentioned, the true intentions of the parties are disregarded in Germany, in Switzerland and in France. The parties are either bound to the simulated contract or also the dissimulated contract is regarded as void. All three legal systems thus operate in contrast to the basic idea underlying the doctrine of simulation in the practically important case of incorrect statements of the price in real estate sales. They thus introduce an element of punishment for fraudulent behaviour into the law of simulation.
4. Simulation in European law
The law of the European Union does not contain general provisions on simulation. The Court of Justice, however, regards the rule that simulated contracts are void as a general principle of European law. The rule has particular importance in connection with the European law regarding subsidies and taxes (see eg ECJ Case C-255/02 – Halifax  ECR I-1609 para 69 on a case of simulation in connection with the European rules on value-added tax). Article 4 para 3 of Reg 2988/95/EC establishes that subsidies have to be returned if they were obtained through acts ‘artificially creating the conditions required for obtaining that advantage’. This provision can be seen as an explicit recognition by the European legislature of the rule that simulated contracts and other legal acts are void.
The restatements and drafts for a unification of European private law contain provisions in accordance with traditional doctrine. Article 6:103 PECL states that, as between the contracting parties, a simulated contract is void while the agreement actually intended is potentially valid. The PECL contain no provision for the protection of third parties. According to the drafters the differences among the European systems on this point do not support a uniform rule.
Article 155 of the draft produced by the Académie des Privatistes Européens (Code européen des contrats (Avant-projet)) contains more detailed rules. In addition to stating the basic rule that simulated contracts are void while dissimulated ones may be valid, the draft provides that a third party may treat the simulated contract as valid or void at his option. The draft follows Italian law in limiting the possibility to call witnesses to prove that a written contract is the product of simulation. This will in many cases result in the parties being bound to the simulated contract.
Article II.-9:201 DCFR contains an extended version of Art 6:103 PECL also dealing with the issue of protection of third parties. Unlike the draft of the Académie des Privatistes, it does not give the affected third party a right to choose how to treat the contract. In fact, a simulated contract remains valid if a third party reasonably relied on it in good faith.
Helmut Coing, ‘Simulatio und Fraus in der Lehre des Bartolus und des Baldus’ in Festschrift Paul Koschaker, vol III (1939, reprinted 1977) 402; Gunter Wesener, ‘Das Scheingeschäft in der spätmittelalterlichen Jurisprudenz, im Usus modernus und im Naturrecht’ in Festschrift Heinz Hübner (1984) 337; Jan Schröder, Gesetzesauslegung und Gesetzesumgehung (1985); Mario Casella, Art ‘Simulazione (dir. priv.)’ in Enciclopedia del diritto, vol XLII (1990) 593; Reinhard Zimmermann, The Law of Obligations (1990), 646 ff; Arzu Oğuz, Probleme der Simulation in rechtshistorischer und rechtsvergleichender Sicht (1998); Andreas Wacke, ‘Mentralreservation und Simulation als antizipierte Konträrakte bei formbedürftigen Geschäften’ in Festschrift Dieter Medicus (1999) 651; Martin Josef Schermaier, ‘§§ 116–124. Willensmängel I’ in Mathias Schmoeckel, Joachim Rückert and Reinhard Zimmermann (eds), Historisch-kritischer Kommentar zum BGB, vol I (2003); Jacques Herbots, ‘La lettre et l’ésprit du contrat dans une perspective de droit comparé: ne point prouver contre ou outre l’écrit’ in Mélanges offerts à Marcel Fontaine (2003) 379; Victor Turonyi, Comparative Tax Law (2003), 157 ff; Alberto Luis Zuppi, ‘The Parol Evidence Rule: A Comparative Study of the Common Law, the Civil Law Tradition, and Lex Mercatoria’  35 Ga J Int’l & Comp L 233.