SUPREME COURT OF CANADA

Appeal heard: December 10, 2015 Judgment rendered: July 29, 2016 Docket: 36205

Between:

Ferme Vi-Ber inc.

Appellant

and

La Financière agricole du Québec

Respondent

And Between:

Simon Cloutier et al.

Appellants

and

La Financière agricole du Québec

Respondent

Official English Translation

Coram: McLachlin C.J. and Abella, Cromwell, Karakatsanis, Wagner, Gascon and Côté JJ.

Joint Reasons for Judgment: (paras. 1 to 95) Reasons Dissenting in Part: (paras. 96 to 136)

Wagner and Gascon JJ. (McLachlin C.J. and Abella, Cromwell and Karakatsanis JJ. concurring) Côté J.

Ferme Vi‑Ber inc. v. Financière agricole du Québec, 2016 SCC 34, [2016] 1 S.C.R. 1032

Ferme Vi‑Ber inc. Appellant

v.

La Financière agricole du Québec Respondent

‑ and ‑

Simon Cloutier, Denis Trépanier, Société coopérative agricole des Bois‑Francs, Coopérative agricole Covilac, Coop Purdel, Société coopérative agricole La Seigneurie, Coopérative agricole Unicoop, 9012‑2151 Québec inc., Sogéporc inc., Gabriel Turgeon inc., Société en commandite Pascoporc, Groupe Dynaco, coopérative agroalimentaire, Coopérative agroalimentaire Comax, R. Rousseau & Fils, S.E.C., Ferme Olympique, S.E.C., Moulées Désy, S.E.C., Inter Agro inc., Techni‑porc inc., Élevage La Bretanne inc., 9038‑7747 Québec inc., Ferme Lor‑re inc., Ginette Marchesseault, Ferme Casimir inc., C&G Paquette inc., Entreprises B. Paquette inc., Porcheries du Button ltée, Élevages du Bas Ste‑Anne inc., Ferme Suporsonique, Gène‑Alliance inc., 9076‑1776 Québec inc., Cultures Excel inc., Francine Sauvageau inc., Ferme Jétizack inc., Cultures Quinto inc., Élevages Hébertville S.E.N.C., Jean‑Marc Henri inc., Ferme Porcéréale inc. (formerly known as Maraîchers de St‑Gilles (1991) S.E.N.C.), Ferme Gosford enr. S.E.N.C., Ferme André Breton inc., Ferme S. & M. Ménard inc., Ferme Luc Loranger inc., Méloporc inc., Ferme Frangis S.E.N.C., Ferme R.M. Côté & Fils (2000) inc., Ferme Porcine Marnie S.E.N.C., F. Ménard inc., Élevages Jacques Joyal inc., Ferme La Ronchonnerie inc., Ferme Mafran inc., Coopérative agricole Profid’Or, Isoporc inc., Ferme Gervais Gosselin inc., R. Robitaille et Fils inc., Groupe CDLM inc., Ferme Gaudreau inc., 9039‑2648 Québec inc., Ferme Denis Robitaille inc., Élevages du Haut‑Richelieu inc., Viaporc inc., Porc S.B. inc., Ferme G. Rompré inc., 9084‑9183 Québec inc., Porcs N&M inc., Ferme Porclair S.E.N.C., Ferme R.D.S. inc., Élevages L.D. ltée, Porc P.G. S.E.N.C., Ferme M.Y. Turgeon inc., Coopérative agricole de St‑Bernard, Élevage Y. Ducharme inc., Production A. Couture (n o 1) ltée, Production A. Couture (n o 2) ltée, Production A. Couture (n o 3) ltée, Production A. Couture (n o 4) ltée, Production A. Couture (n o 5) ltée, Production A. Couture (n o 6) ltée, Production René Lait inc., Alfred Couture Limitée, Ferme B.E.L. Porcs ltée, Porcs M.L. inc., Ferme Vallières & Gosselin inc., Meunerie St‑Elzéar ltée, Élevages Labrecque inc., Joly‑Grains inc., Joly‑Porcs inc., Site de la Colline inc., Site des Érables inc., Ferme Serge inc., Ferme Jolivoir inc., Aliments Breton inc., Ferme C.B. inc., Fermili inc., Luma Genetic Inc. (formerly known as Génétiporc inc.), Entreprises Magnum inc., Trans‑Porcité inc., Lait‑Porcité inc., Ferme C.M. S.E.N.C., Ferme Porc Saint S.E.N.C., Entreprises Rémy Laterreur inc., Rémy Laterreur, Élevages Explorateurs inc., Ferme Palene inc., Ferme André Hénault S.E.N.C., Germain Lapointe, Ferme Jenlica inc., Immeubles Clément Dubois inc., Fermes Roda inc., Fermes Richard inc., Ferme Jocko S.E.N.C., Ferme D.J. Frappier inc., Entreprises Paul Claessens inc., Ferme H. et M. Potvin S.E.N.C., Ferme Jean‑Paul Palardy inc., Entreprises Denis Lacoste inc., Chantal D’Amour, Ferme Bonneterre inc., Ferme D’Anjou & Fils inc., M.B.M. Daigle S.E.N.C., Ferme Réjean Turgeon inc., Ferme Jymdom inc., Ferme Jules Côté et Fils inc., Ferme D.M.L. inc., Ranch St‑Sylvestre inc., John Houley inc., Ferme Belgica inc., Ferme Bovipro S.E.N.C., Jacques Desrosiers, Éric Desrosiers, Ferme B&L Desrosiers S.E.N.C., 9078‑1170 Québec inc., Fermes St‑Henri, S.E.C., Ferme Ray‑Loi, S.E.C., Fermes St‑Apollinaire, S.E.C., Élevages St‑Félix, S.E.C., Élevages St‑Patrice, S.E.C. and Ferme Beaumontoise, S.E.C. Appellants

v.

La Financière agricole du Québec Respondent

Indexed as: Ferme Vi‑Ber inc. v. Financière agricole du Québec

2016 SCC 34

File No.: 36205.

2015: December 10; 2016: July 29.

Present: McLachlin C.J. and Abella, Cromwell, Karakatsanis, Wagner, Gascon and Côté JJ.

on appeal from the court of appeal for quebec

Agriculture — Farm income stabilization — Compensation — Calculation method — Legal framework applicable to Quebec’s Programme d’assurance stabilisation des revenus agricoles — Participants in program contesting method for calculating compensation payments that was adopted by La Financière agricole du Québec to take federal government grants to farmers into account — Whether program is contract and, if so, whether it is subject to rules applicable to contract of insurance within meaning of Civil Code of Québec — Whether La Financière, in determining compensation payable to participants under program, acted in conformity with its rights and obligations by linking amounts at issue collectively — Programme d’assurance stabilisation des revenus agricoles, 2001, 133 G.O. 1, 1336, s. 88(3).

La Financière agricole du Québec (“La Financière”) is a legal person established in the public interest under the Act respecting La Financière agricole du Québec . Its mission is to support and encourage the development of the agricultural and agro‑food sector within the perspective of sustainable development. For that purpose, it has set up income protection, insurance and farm financing programs. The appellants are Quebec farm producers that participated voluntarily in the Programme d’assurance stabilisation des revenus agricoles (“ASRA Program”) administered by La Financière. The ASRA Program protects participants from having their income drop below a level defined by La Financière for 10 agricultural products or classes of products designated as “insurable”. That level is reached where the “net annual income” of an average benchmark farm for an insured product is less than the “stabilized net annual income”, which corresponds to a percentage of the average annual regular salary of a skilled worker in Quebec. In short, the purpose of the ASRA Program is to guarantee that an average farm producer never earns less than a predetermined percentage of the average income of a skilled worker. Each producer participating in this voluntary program must pay a fixed contribution per unit of a designated product, agree to participate for a minimum of five years and insure all of their annual production for each designated product. La Financière makes a contribution to the program’s fund — the Fonds d’assurance stabilisation des revenus agricoles, of which it is the trustee — equal to twice the contributions paid by each participant. The amounts in the fund are used to finance the payment of compensation to participants.

The appellants contested certain decisions made by La Financière in determining their compensation payments for 2007. Those decisions were related to the calculation method chosen by La Financière, in determining the compensation payable under the program, to take account of additional income received as farm financial assistance from the federal government. Both the parties and the courts below used the word “linkage” to characterize the process of taking such income into account, which is provided for in s. 88(3) of the ASRA Program. Amounts so received are linked either “collectively” — on the basis of the amounts the average benchmark farm would have received — or “individually” — on the basis of the amounts each ASRA Program participant actually received from the various governments. La Financière deducts the amounts so “linked” from the compensation. The appellants argued that the ASRA Program was a contract of insurance and that La Financière had, by collectively linking certain amounts received as financial assistance, improperly incorporated additional income into its calculations so as to reduce their compensation under the program in violation of the terms of the contract, which had to be interpreted on the basis of their reasonable expectations as insured persons.

The appellants applied to the Superior Court, which allowed their action, characterizing the ASRA Program as a contract of insurance and ordering La Financière to pay them substantial additional compensation for 2007. The Court of Appeal set aside that judgment, finding that the ASRA Program was not a contract of insurance and that the impugned decisions were reasonable.

Held (Côté J. dissenting in part): The appeal should be dismissed.

Per McLachlin C.J. and Abella, Cromwell, Karakatsanis, Wagner and Gascon JJ.: Despite the broad discretion conferred on La Financière by the Act respecting La Financière agricole du Québec and the ASRA Program, the program cannot be considered simply a government program that is governed by public law. As can be seen from a review of its structure and how it functions, it is different from two classic examples of social programs that fall under public law: social insurance programs and agricultural subsidies. Unlike a social insurance program, it applies to only one sector and is neither universal nor compulsory, and its benefits are not calculated using simple formulas applicable to broad classes of persons and situations. It includes several contract‑style clauses for terminating the contract for predetermined reasons, which create acquired rights for the current year. Moreover, La Financière’s considerable management autonomy is limited by the need to comply with the contractual conditions that bind it and the participants. These characteristics, together with the contributions required from participants, also distinguish the ASRA Program from simple agricultural subsidy programs granted on an ex gratia basis and without consideration.

This program in fact has the characteristics of an administrative contract, that is , a contract to which a public authority is a party, and all the rules needed to guide the actions of the parties can be found in private law. However, administrative contracts are distinguishable from contracts between private parties, since parity between the parties does not always exist. Where the government’s contractual relations are concerned, therefore, the public interest must be considered in interpreting those relations and may weigh in favour of a broader discretion in implementing the government scheme, especially where that scheme has a social objective. These are not principles of public law, but considerations related to the object of the contract that may influence the interpretation of the scope of the contractual powers of the public authority in question. The government’s discretion nonetheless has its limits . In the context of an administrative contract, those limits do not derive from the public law duty of procedural fairness but are instead based on good faith and contractual fairness, which flow, in Quebec law, from the application of arts. 6, 7, 1375 and 1434 of the Civil Code of Québec (“ C.C.Q. ”).

Furthermore, the ASRA Program is an innominate administrative contract that does not have the three main characteristics of a contract of insurance set out in art. 2389 C.C.Q. , namely (i) an obligation on the client to pay a premium or assessment; (ii) the occurrence of a risk; and (iii) an obligation on the insurer to make a payment to the client if the insured risk occurs. It cannot therefore be subject to the rule of interpretation based on the reasonable expectations of the insured that applies to a contract of insurance and, in Quebec law, applies solely in its minimum dimension, that is, to resolve any ambiguity in the terms of the contract in favour of the insured.

For the purpose of determining whether it was open to La Financière to collectively link the amounts received under the federal assistance programs at issue in this case, it is the rules of contractual interpretation set out in arts. 1425 to 1432 C.C.Q. that apply. Section 88(3) of the ASRA Program does not specify how compensation received under government assistance programs is to be linked. It states only that, for this purpose, La Financière must consider “[a]ny amounts to which a participant is entitled” under such programs. When properly interpreted in light of the contract as a whole (art. 1427 C.C.Q .) and La Financière’s past practices (art. 1426 C.C.Q .), s. 88(3) of the ASRA Program does not require that amounts be linked individually but, on the contrary, gives La Financière the discretion to decide what linkage method to employ.

The general structure of the program and the contract as a whole support the conclusion that collective linkage is normally required. In fact, a reading of s. 88(3) of the ASRA Program in the context of the program as a whole, which is based on the collective concept of a benchmark farm, leads to the conclusion that collective linkage must be preferred. Section 88(3) is in Division XI, the title of which refers to the collective concept of “Farm Models”. Moreover, ss. 86 and 92 of the ASRA Program clearly state that the net annual income used to calculate the compensation, which includes amounts received from government sources, is that of a “specialized benchmark farm for each of the products or classes of products”.

Although La Financière has sometimes linked amounts individually in the past, it appears from the evidence that decisions to do so were usually based on the number of participants that received the amounts in question and not on the fact that government assistance had been paid directly to producers. As a result, neither the contract as a whole nor past practice supports a conclusion that La Financière was under a statutory or contractual obligation to link the amounts individually in this case.

Although La Financière had the discretion to link the amounts collectively, it was required to exercise that discretion in accordance with the requirements of good faith and contractual fairness. The decision to link the amounts collectively in this case was made following extensive consultations with representatives of farm producers and after impact simulation studies had shown that most of the program’s participants would benefit from that decision. By linking them collectively, La Financière also favoured the smallest producers. This situation was consistent with La Financière’s mission. Finally, the decision to link the amounts collectively was compatible with the specific features of the federal programs at issue. La Financière thus exercised its powers in accordance with the requirements of good faith and contractual fairness. It was open to La Financière in fixing the compensation payable to the appellants to choose to link the amounts they had received under the relevant federal financial assistance programs collectively, which means that the appellants are not entitled to the amounts they claim.

Per Côté J. (dissenting in part): The only determinative issue in this case is one of contractual interpretation. Regardless of whether the contract is characterized as a contract of insurance or as an innominate contract that falls under both public law and private law, the result is the same. It is true that the appellants participated voluntarily in the ASRA Program, but insofar as that program involves a contract of adhesion imposed by La Financière, that is, a contract that is not negotiated with participants, the characterization makes no difference. If there is any ambiguity, it must be resolved in favour of the adhering party in accordance with art. 1432 C.C.Q . The rule of interpretation of reasonable expectations adds nothing to the existing rules of interpretation.

The reason why producers participate in the ASRA Program is simple: they expect to receive the full compensation they are owed in return for paying their contributions. When La Financière deprives them of all or part of the compensation to which they are entitled, judicial intervention is warranted. The common intention of the parties, the overall scheme of the ASRA Program and past practices confirm that La Financière contravened the program in deciding to subtract excessively high amounts of notional income from the amounts to which participants were otherwise entitled.

It sometimes happens that during the year individual participants receive additional amounts to which they are personally entitled from other government agencies. The program expressly authorizes La Financière to take such amounts into account in its calculations to ensure that participants do not receive double compensation for a single loss. However, although s. 88(3) of the ASRA Program does authorize La Financière to ensure that participants do not receive double compensation, it does not permit La Financière to attribute notional amounts to some participants in order to overcompensate other participants for policy reasons.

In this case, not only did the chosen linkage method not make it possible to avoid double compensation, it also prevented many participants from receiving the full compensation to which they were entitled under the ASRA Program. While it is true that La Financière’s mission is broad, that mission does not authorize it to subvert the purpose of s. 88(3) by assuming a discretion it does not have.

Once the ASRA Program had been adopted, La Financière had to comply with the rules of the game it had itself established. In the case of amounts received directly from other granting organizations that can be considered to be annual receipts, s. 88(3) provides that, in calculating annual receipts, La Financière must take into account “[a]ny amounts to which a participant is entitled on the basis of the volume of marketed products and secondary products”. La Financière may not therefore penalize a participant for amounts to which he or she is not entitled owing to the inherent limits of those other programs. By referring to a “participant”, s. 88(3) requires La Financière to take into account the amounts that were actually received. The definition set out in s. 2 refers to an individual participant in the ASRA Program, not to a “benchmark farm”, and the same is true of the definition of “annual receipts”. If La Financière disregards the inherent limits of those programs and if the amounts attributed to participants bear no relation to the amounts they actually received, it is in breach of its contractual obligations.

The evidence shows that, where amounts have been granted directly to a producer under another program in the past, La Financière has never used a linkage method — whether collective or individual — that had the effect of negating the compensation to which participants were entitled by attributing amounts that disregarded the inherent limits of those programs. La Financière’s choice in the cases at issue is therefore also inconsistent with its past practices.

What is problematic in this case is that La Financière’s action had the effect of overcompensating certain participants to the detriment of the others, that it disregarded the impact of the caps under the federal programs on the insurance coverage of participants and that the amounts it ultimately attributed bore no relation to the amounts that had actually been received, not that it chose a particular linkage method. This was the conclusion reached by the trial judge, and there is no reason to intervene in this regard.

The producers’ appeal should be allowed, but the trial judge’s award should be reduced by subtracting from it an amount equal to the contributions the producers would have had to pay in exchange for higher compensation.

Cases Cited

By Wagner and Gascon JJ.

Referred to: Law v. Canada (Minister of Employment and Immigration) , 1999 675 (SCC) , [1999] 1 S.C.R. 497; Peters v. Canada (Attorney General) , 2009 FC 400 ; Martin Service Station Ltd. v. Minister of National Revenue , 1976 208 (SCC) , [1977] 2 S.C.R. 996; Confédération des syndicats nationaux v. Canada (Attorney General) , 2008 SCC 68 , [2008] 3 S.C.R. 511; Trépanier v. Financière agricole du Québec , 2011 QCCS 1802 ; Jacobs v. Agricultural Stabilization Board , 1982 157 (SCC) , [1982] 1 S.C.R. 125; George A. Demeyere Tobacco Farms Ltd. v. Continental Insurance Co. (1984), 1984 1823 (ON SC) , 46 O.R. (2d) 423; Brissette v. Financière agricole , 2006 QCCS 1620 ; Canada (Attorney General) v. Mavi , 2011 SCC 30 , [2011] 2 S.C.R. 504; Glykis v. Hydro‑Québec , 2004 SCC 60 , [2004] 3 S.C.R. 285; Financière agricole du Québec v. Forand , 2009 QCCQ 10263 ; Martel Building Ltd. v. Canada , 2000 SCC 60 , [2000] 2 S.C.R. 860; Colautti Brothers Marble Tile & Carpet (1985) Inc. v. Windsor (City) (1996), 36 M.P.L.R. (2d) 258 ; Rollo Bay Holdings Ltd. v. Prince Edward Island Agricultural Development Corp. (1993), 1993 1777 (PE SCAD) , 110 D.L.R. (4th) 132; Financière agricole du Québec v. Coddington , 2013 QCCQ 6238 ; Lafortune v. Financière agricole du Québec , 2016 SCC 35 , [2016] 1 S.C.R. 1091; Brissette Estate v. Westbury Life Insurance Co. , 1992 32 (SCC) , [1992] 3 S.C.R. 87; Jesuit Fathers of Upper Canada v. Guardian Insurance Co. of Canada , 2006 SCC 21 , [2006] 1 S.C.R. 744; Reid Crowther & Partners Ltd. v. Simcoe & Erie General Insurance Co. , 1993 150 (SCC) , [1993] 1 S.C.R. 252; National Bank of Greece (Canada) v. Katsikonouris , 1990 92 (SCC) , [1990] 2 S.C.R. 1029; Excellence (L’), compagnie d’assurance‑vie v. Desjardins , 2005 QCCA 1035 , [2005] R.R.A. 1085; Affiliated FM Insurance Co. v. Hafner Inc. , 2006 QCCA 465 ; Souscripteurs du Lloyd’s v. Alimentation Denis & Mario Guillemette inc. , 2012 QCCA 1376 ; Consolidated‑Bathurst Export Ltd. v. Mutual Boiler and Machinery Insurance Co. , 1979 10 (SCC) , [1980] 1 S.C.R. 888; Industrielle, Compagnie d’Assurance sur la Vie v. Bolduc , 1978 171 (SCC) , [1979] 1 S.C.R. 481.

Statutes and Regulations Cited

Act respecting La Financière agricole du Québec , CQLR, c. L‑0.1, ss. 1, 3 , 19 , 22 .

Agricultural Products Insurance Act, 1996 , S.O. 1996, c. 17, sch. C.

Agricultural Stabilization Act , R.S.C. 1970, c. A‑9.

Canada Pension Plan , R.S.C. 1985, c. C‑8.

Civil Code of Québec , arts. 6, 7, 1375, 1425 to 1432, 1426, 1427, 1434, 2389, 2408 to 2413, 2466 to 2468, 2470 to 2474.

Employment Insurance Act , S.C. 1996, c. 23.

Immigration and Refugee Protection Act , S.C. 2001, c. 27.

Programme d’assurance récolte , (2002) 134 G.O. 1, 261, ss. 10, 15, 27, 35 to 37, 38, 42.

Programme d’assurance stabilisation des revenus agricoles , (2001) 133 G.O. 1, 1336, ss. 1, 2 “ adhérent ”, “ recettes annuelles ”, 6, 7, 13, 16(3), 18, 19, 21, 22, 78, 80, 86, 87, 88 [am. (2009) 141 G.O. 1, 51, s. 21], 89, 92, 101, 103.

Unemployment Insurance Act, 1971 , S.C. 1970‑71‑72, c. 48 [repl. 1996, c. 23].

Authors Cited

Bergeron, Jean‑Guy. Les contrats d’assurance (terrestre): lignes et entre-lignes , t. 1. Sherbrooke: SEM Inc., 1989.

Dussault, René, and Louis Borgeat. Administrative Law: A Treatise , vol. 1, 2nd ed. Translated by Murray Rankin. Toronto: Carswell, 1985.

Garant, Patrice, avec la collaboration de Philippe Garant et Jérôme Garant. Droit administratif , 6 e éd. Cowansville, Que.: Yvon Blais, 2010.

Issalys, Pierre, et Denis Lemieux. L’action gouvernementale: Précis de droit des institutions administratives , 3 e éd. Cowansville, Que.: Yvon Blais, 2009.

Lluelles, Didier. Précis des assurances terrestres , 5 e éd. Montréal: Thémis, 2009.

Lluelles, Didier, et Benoît Moore. Droit des obligations , 2 e éd. Montréal: Thémis, 2012.

Thouin, Marie‑Chantal. “La théorie de l’attente raisonnable de l’assuré” (1997), 64 Assurances 545.

APPEAL from a judgment of the Quebec Court of Appeal (Hilton, Gagnon and Savard JJ.A.), 2014 QCCA 1886 , [2014] AZ‑51115390, [2014] J.Q. n o 11218 (QL), setting aside a decision of Monast J., 2012 QCCS 284 , [2012] AZ‑50827524, [2012] J.Q. n o 701 (QL), 2012 CarswellQue 666 (WL Can.). Appeal dismissed, Côté J. dissenting in part.

Bruno Lepage , Madeleine Lemieux and Dominique‑Anne Roy , for the appellants.

Matthieu Brassard , Jean‑Pierre Émond and Valérie Blanchet , for the respondent.

English version of the judgment of McLachlin C.J. and Abella, Cromwell, Karakatsanis, Wagner and Gascon JJ. delivered by

Wagner and Gascon JJ. —

I. Overview

[ 1 ] At issue in this appeal is the application of private law to the implementation of certain government financial support programs. More specifically, the Court must identify the rules governing the interpretation of the rights and obligations of the parties to the Programme d’assurance stabilisation des revenus agricoles , (2001) 133 G.O. 1, 1336 (“ASRA Program”), administered by the respondent, La Financière agricole du Québec (“La Financière”).

[ 2 ] The appellants are Quebec farm producers that participated voluntarily in the ASRA Program. Under that program, La Financière undertook, in return for contributions from producers, to protect them from the income fluctuations associated with the agricultural market. The appellants contested certain decisions made by La Financière in determining their compensation payments for 2007. Those decisions were related to the calculation method chosen by La Financière, in determining the compensation payable under the program, to take account of additional income received as farm financial assistance from the federal government. The appellants argued that the ASRA Program was a contract of insurance and that La Financière had improperly incorporated that additional income into its calculations so as to reduce their compensation under the program, in violation of the terms of the contract, which had to be interpreted on the basis of their reasonable expectations as insured persons.

[ 3 ] The producers applied to the Superior Court, which allowed their action, characterizing the ASRA Program as a contract of insurance and ordering La Financière to pay them substantial additional compensation for 2007. The Court of Appeal set aside that judgment, finding that the ASRA Program was not a contract of insurance and that the impugned decisions were reasonable.

[ 4 ] We would dismiss the appeal. The ASRA Program is not a contract of insurance but simply an innominate contract under the civil law. It cannot be subject to the rule of interpretation based on the reasonable expectations of the insured that applies to a contract of insurance as defined in the Civil Code of Québec (“ C.C.Q. ”). While it is true that the contract must be interpreted having regard to the public interest and to La Financière’s social objective, it is nonetheless governed exclusively by private law, not by public law. For the purpose of determining the compensation payable to its participants, the ASRA Program gives La Financière a discretion to determine how to calculate any other income they have received from government sources. La Financière exercised that discretion in accordance with the requirements of good faith and contractual fairness, which means that the appellants are not entitled to the amounts they claim.

II. Background

[ 5 ] La Financière is a legal person established in the public interest under the Act respecting La Financière agricole du Québec , CQLR, c. L‑0.1 (“ AFAQ ”). Its mission is to “support and encourage the development of the agricultural and agro‑food sector within the perspective of sustainable development” ( s. 3 AFAQ ). For that purpose, it has set up income protection, insurance and farm financing programs.

[ 6 ] Among other things, La Financière offers a crop insurance service that compensates participants for losses caused, for example, by unfavourable and unpredictable weather conditions. This Programme d’assurance récolte , which is published in the Gazette officielle du Québec , (2002) 134 G.O. 1, 261, has equivalents in several other Canadian provinces. In Ontario, for instance, there is the Agricultural Products Insurance Act, 1996 , S.O. 1996, c. 17, sch. C. La Financière also offers the ASRA Program, which provides financial support to producers. This appeal concerns the interpretation of the rights and obligations of the parties to the ASRA Program, which had no equivalent elsewhere in Canada at the time the dispute in this case arose. The 137 appellants are farm producers that participated voluntarily in the program. The ASRA Program is published in the Gazette officielle du Québec . It replaced the Farm Income Stabilization Insurance Scheme in 2001; the latter had been created in 1975 and was administered first by the Commission administrative des régimes d’assurance‑stabilisation des revenus agricoles until 1979, and then by the Régie des assurances agricoles du Québec until 2001, the year La Financière was established ( s. 1 AFAQ ).

[ 7 ] The ASRA Program protects participants from having their income drop below a level defined by La Financière for 10 agricultural products or classes of products designated as [ translation ] “insurable”. That level is reached where the “net annual income” of an average benchmark farm for an insured product is less than the “stabilized net annual income”, which corresponds to a percentage of the average annual regular salary of a skilled worker in Quebec (s. 89 ASRA Program). What this means is that the higher the net annual income of the benchmark farm is, the lower the compensation will be. In essence, the purpose of the program is to guarantee that an average farm producer never earns less than a predetermined percentage of the average income of a skilled worker.

[ 8 ] In exchange, each producer participating in this voluntary program must pay a fixed contribution per unit of a designated product (s. 78 ASRA Program). Producers must agree to participate for a minimum of five years (s. 16(3)) and must insure all of their annual production for each designated product (s. 18). La Financière makes a contribution to the program’s fund — the Fonds d’assurance stabilisation des revenus agricoles, of which it is the trustee (s. 6) — equal to twice the contributions paid by each participant (s. 80). The amounts in the fund are used to finance the payment of compensation to participants (s. 7), and, unless an agreement is entered into or an alternative program is implemented, any surplus or deficit must be apportioned among the participants in proportion to their contributions (s. 13). No amendment made by La Financière to the terms of the ASRA Program, except one relating to the contribution rate, may take effect until the insurance year after the amendment comes into force (s. 21 para. 2).

[ 9 ] The first paragraph of s. 87 of the program provides that [ translation ] “[t]he net annual income [of a benchmark farm] corresponds to the annual receipts minus cash disbursements and depreciation.” The annual receipts of a benchmark farm are described in s. 88 of the ASRA Program. Section 88(3), as it read at the time this dispute arose, read as follows concerning amounts granted by government agencies:

[ translation ]

3° Any amounts to which a participant is entitled on the basis of the volume of marketed products and secondary products and that are granted by government agencies in the form of price compensation for the insurable product or under a government farm business risk management program.

Thus, in calculating the annual receipts of the average benchmark farm, La Financière takes account of other income from federal and provincial government contributions under a risk management program or in the form of price compensation for insurable products. The parties use — as did the courts below — the word [ translation ] “linkage” to characterize the process of taking such income into account. From the compensation to be paid under the ASRA Program, La Financière therefore deducts the “linked” amounts in order to take account of any other income that might have an effect on the financial needs of the average benchmark farm, which is the reference standard for the program.

[ 10 ] Amounts so received are linked either “collectively” — on the basis of the amounts the average benchmark farm would have received — or “individually” — on the basis of the amounts each ASRA Program participant actually received from the various governments. The higher the linked amounts, the lower the compensation received under the program, since increases in the benchmark farm’s income that result from the linkage process translate directly into a decrease in the compensation paid to each participant under the program. Whether government assistance should be linked collectively or individually is at the heart of the dispute between the appellants and La Financière.

[ 11 ] La Financière often acts as an intermediary in administering federal or provincial grants to Quebec farm producers. For that purpose, it has entered into a number of agreements with the federal government to distribute amounts that the federal government pays directly to producers in other provinces. At times, La Financière also integrates amounts paid by the federal government directly into the ASRA Program’s fund.

[ 12 ] In May and July 2007, the federal government announced the payment of grants to Canadian farmers (including ASRA Program participants) under two programs: the Cost of Production Benefit (“COPB”) program and the AgriInvest Kickstart (“Kickstart”) program. Under these two assistance programs, grants were made to producers, who were not required to make contributions.

[ 13 ] Amounts paid under the two programs were calculated as a percentage of each producer’s allowable net sales (“ANS”), although both programs had ANS caps above which producers received nothing. The caps were $450,000 for the COPB program and $3 million for Kickstart. They corresponded to maximum payments per producer of $12,240 for the COPB program and $96,000 for Kickstart. All ASRA Program participants were eligible to receive amounts under those programs, but the largest producers received only the maximum amounts.

[ 14 ] In managing the ASRA Program, La Financière decided to link amounts received under the COPB and Kickstart programs collectively. It fixed the compensation for 2007 on the basis of what the benchmark farm would have received under those two programs. The ANS of the hypothetical benchmark farm used to collectively link the amounts were less than the maximum allowed by the federal government. However, in collectively linking the amounts, La Financière calculated the average income the benchmark farm would have earned per unit of designated product and then multiplied this by each participant’s number of units of the designated product. These calculations led La Financière to attribute to some of the largest producers, including the appellants, amounts of federal assistance greater than the amounts they had actually received. As a result, the compensation paid to these producers under the ASRA Program was reduced by more than would have been the case had La Financière linked the amounts individually on the basis of what each of them had actually received.

[ 15 ] The 137 appellants were unhappy about being disadvantaged in this way, and in January 2008 they asked La Financière to review its decision and link the amounts in question individually. Their request was denied on the ground that a review could not be sought for decisions relating to the [ translation ] “terms of the programs administered by La Financière”.

[ 16 ] In November 2008, La Financière amended the ASRA Program to specify that amounts received under other financial assistance programs would now be linked collectively [ translation ] “unless La Financière agricole considers it appropriate” to link them individually (s. 88(3) added by (2009) 141 G.O. 1, 51, s. 21).

[ 17 ] Following a fruitless exchange of letters with La Financière, the appellants instituted proceedings in the Superior Court in two cases that were ultimately joined for hearing. They argued that the ASRA Program had to be characterized as a contract of insurance within the meaning of the Civil Code of Québec . In their submission, this characterization engaged the principle of interpretation based on the [ translation ] “reasonable expectations of the insured”, sometimes also called the doctrine of “legitimate expectations of the insured”, according to which any ambiguity, or even any unambiguous provision of a contract of insurance, must be interpreted in a manner consistent with the expectations of the insured. They submitted that, in light of past practice, it was reasonable for them to expect La Financière to link the amounts paid by the federal government individually. The decision to do so collectively was therefore, in their view, [ translation ] “arbitrary, discriminatory and improper” because of its negative effect on the compensation paid to many participants. That decision resulted in the amendment of the contract in the course of an insurance year, which the program did not permit, and deprived the appellants of amounts to which they were entitled.

[ 18 ] In their conclusions, the appellants asked the court to declare that La Financière [ translation ] “must, under the ASRA Program, deduct any amounts to which each participant is entitled under a federal program on an individual basis”. They also asked that the calculation that resulted from linking the COPB and Kickstart amounts be declared to be invalid in respect of all ASRA Program participants (who numbered 16,747 at the time). Each of them claimed the amount it would have received for 2007 had the amounts been linked individually, a total of over $14 million.

[ 19 ] La Financière countered that the ASRA Program is not a contract of insurance, adding that the AFAQ gives it a broad discretion to determine the amount of the assistance granted to producers and to establish the conditions applicable to such assistance, including the method to be employed in linking the amounts it must take into account in granting compensation. Linking them collectively was in its view consistent with the nature of the ASRA Program, which is based on a benchmark farm and not on data specific to each participant. La Financière also argued that the chosen method had been advantageous to a clear majority of participating producers. It explained that it had in the past linked amounts individually only in cases in which a single group of ASRA Program participants had received amounts to be linked, and that it had done so in such cases to avoid attributing income to all the participants that some of them had not received. La Financière added that it had exercised its power to decide on the method to employ in linking amounts on a case‑by‑case basis having regard to its mission and to the impact of each contribution on the participants as a whole.

III. Judicial History

A. Superior Court, 2012 QCCS 284

[ 20 ] The trial judge found that the ASRA Program was a contract of insurance, explaining that La Financière insured a risk in exchange for the payment of a contribution. She noted that La Financière had the power to decide what data would be taken into account in calculating the compensation that was payable, but that any unilateral amendment of the terms of the ASRA Program by La Financière could apply only to the year following the amendment and only after having been published in the Gazette officielle du Québec .

[ 21 ] The trial judge observed that, before being amended in November 2008, s. 88(3) had referred expressly to [ translation ] “amounts to which a participant is entitled”. There was therefore nothing preventing La Financière from linking the amounts individually in this case. The November 2008 amendments required that La Financière determine the amounts to which a participant was entitled on the basis of the characteristics of the benchmark farm (collective linkage) unless La Financière considered it appropriate to add the amounts actually received to the annual receipts (individual linkage). In the trial judge’s view, that amendment [ translation ] “suggests that the intention was to change either the law or the parties’ status” (para. 157 ()).

[ 22 ] Further, the trial judge concluded on the basis of the evidence that, in the past, La Financière had collectively linked amounts paid under programs involving [ translation ] “a single contribution paid directly to its insurance fund [that of the ASRA Program]”, but had linked the amounts of “direct payments to producers” individually by subtracting from the compensation payable to each participant any amounts it had received from the federal government (paras. 133‑34). The financial assistance provided under the two federal programs at issue took the form of direct payments to producers. La Financière’s past practice had therefore created “legitimate expectations among the participants”, which means that La Financière could not in the course of a year decide to change the applicable rules and collectively link amounts paid directly to producers (para. 158).

[ 23 ] The trial judge found that La Financière had a “discretion” to decide what method to employ in linking amounts, but that this discretion had to be exercised [ translation ] “in accordance with the terms of the program . . . and in a reasonable manner”, which was not the case here (paras. 161‑62). By collectively linking the amounts in issue, La Financière had “indirectly reduced the insurance coverage” to a significant degree (para. 164). The extent of the resulting distortions and the fact that the appellants were accordingly penalized led the trial judge to conclude that La Financière had exercised its discretion unreasonably and had acted in an arbitrary, discriminatory and improper manner.

[ 24 ] The trial judge therefore allowed the action, declared the calculation of the compensation paid for 2007 — but only to the appellants — to be invalid, and ordered La Financière to pay the compensation the appellants would have received had the amounts been linked individually, with costs.

B. Court of Appeal, 2014 QCCA 1886

[ 25 ] Savard J.A., writing for the Court of Appeal, began her analysis by observing that participation in the ASRA Program was voluntary and that the program had all the characteristics of a contract. However, she was of the view that the program was not a contract of insurance within the meaning of art. 2389 C.C.Q. She noted, inter alia , that La Financière’s primary activity was not speculating on risk but promoting the agricultural and agri‑food sector. The economic risk covered by the ASRA Program did not depend on each participant’s financial situation, but was always present and was certain over a long period of time (para. 69 ()). In Savard J.A.’s view, the ASRA Program was more of an income protection program, as the government provided farm producers with financial support to counter losses that were inherent in their sector. She described the program as a sui generis administrative contract that had both public law and private law aspects.

[ 26 ] In Savard J.A.’s opinion, s. 88(3) of the ASRA Program did not require La Financière to link amounts individually, but authorized it to take the specific features of each program into account in deciding what calculation method to use. The fact that the words [ translation ] “a participant” had been used in that provision at the relevant time was not conclusive. The purpose of the ASRA Program, which complemented other government assistance programs, meant that La Financière had to be able to determine on a case‑by‑case basis how best to act in the interest of all participants.

[ 27 ] Savard J.A. then analyzed the reasonableness of La Financière’s decision. She found that the trial judge had erred in essentially basing her analysis on the impact of the chosen linkage method on the participants who were disadvantaged by it, and in not considering the purpose of the AFAQ and the interests of all the participants. The decision to link the amounts collectively was not unreasonable. It was based in part on the fact that the federal programs applied both to products that were insured by the ASRA Program and to products that the ASRA Program did not insure. Linking the amounts collectively made it possible to take only payments related to insured products into account. Another reason why La Financière took a collective approach here was to avoid negating the caps established under the federal assistance programs, which was a factor it could include in its analysis. As well, unlike the programs for which amounts had been linked individually in the past, the two federal programs at issue here involved the payment of amounts to every ASRA Program participant. All these characteristics were taken into account in the decision to link the amounts collectively in the circumstances of this case.

[ 28 ] In light of this conclusion, Savard J.A. did not address La Financière’s argument that the action could not be allowed without declaring the linkage calculation to be invalid in respect of all participants. The court therefore allowed the appeal, without costs in view of the parties’ relationship and the nature of the questions that had been raised.

IV. Issue

[ 29 ] The central issue in this appeal is whether La Financière, in determining the compensation payable to the appellants under the ASRA Program for 2007, acted in conformity with its rights and obligations by linking the amounts in question collectively. To decide this issue, it will be essential to begin by outlining the legal framework applicable to the ASRA Program and determining whether the program constitutes a contract and, if so, whether it is a contract of insurance within the meaning of the Civil Code of Québec . We will consider this last subject first. Our analysis in that regard will enable us to define the legal framework applicable to the ASRA Program and identify the nature of the relationship between La Financière and the participants.

V. Analysis

A. Is the ASRA Program a Contract and, if So, Is It Subject to the Rules Applicable to Contracts of Insurance?

(1) Is the ASRA Program a Contract?

[ 30 ] To decide whether the rules of private law or public law apply to the ASRA Program, we must first determine whether the program is a contract.

[ 31 ] Despite the broad discretion conferred on La Financière by the legislation and the ASRA Program (s. 19 AFAQ and s. 87 para. 3 ASRA Program), and even though La Financière finances the compensation paid under the program partly out of public funds (s. 80), the program cannot be considered simply a government program that is governed by public law. The ASRA Program has several features that justify considering it to be a contract. This is clear from a review of its structure and how it functions, which differentiate it from two classic examples of social programs that fall under public law: “social insurance” programs and agricultural subsidies.

[ 32 ] It is true that a government program may include the [ translation ] “basic elements of a contract of insurance” but not its “contractual form”, so that it is in reality a “social insurance” program: P. Issalys and D. Lemieux, L’action gouvernementale: Précis de droit des institutions administratives (3rd ed. 2009), at p. 819; R. Dussault and L. Borgeat, Administrative Law: A Treatise (2nd ed. 1985), vol. 1, at pp. 135‑36. Professors Issalys and Lemieux define the social insurance program as follows:

[ translation ] The early years of modern social security were marked by the creation of social insurance mechanisms. According to this technique, the elements of the legal relationship of insurance, as set out in article 2389 of the Civil Code of Québec , were adopted, but modified to serve the needs of social policy. It can therefore be defined as a statutory mechanism under which a class of persons subject to the mechanism are required to pay premiums to an insurer vested with public authority in exchange for the payment of benefits under the mechanism to those persons or to other recipients on the occurrence of a specified risk. [p. 818]

[ 33 ] It is also well established that a true social insurance program is not contractual in nature. For example, the Canada Pension Plan , R.S.C. 1985, c. C‑8, was described by this Court as “compulsory social insurance” in Law v. Canada (Minister of Employment and Immigration) , 1999 675 (SCC) , [1999] 1 S.C.R. 497, at para. 8 . That plan is governed by public law in the sense that it is managed by an administrative tribunal and that the recourse available to a dissatisfied person to whom the plan applies is to seek judicial review of the tribunal’s decision: see, for example, Peters v. Canada (Attorney General) , 2009 FC 400 . The same is true of the scheme established by the Unemployment Insurance Act, 1971 , S.C. 1970‑71‑72, c. 48 (since replaced by the Employment Insurance Act , S.C. 1996, c. 23), which this Court characterized as “a scheme of compulsory public insurance which was never expected to function on a strict actuarial basis”: Martin Service Station Ltd. v. Minister of National Revenue , 1976 208 (SCC) , [1977] 2 S.C.R. 996, at p. 1005. The Court has also legally characterized the employment insurance premium as a “regulatory charge”, which it defined as “a form of special levy connected with a government program”: Confédération des syndicats nationaux v. Canada (Attorney General) , 2008 SCC 68 , [2008] 3 S.C.R. 511, at paras. 71‑72 .

[ 34 ] Social insurance programs generally share the following characteristics: they apply to more than one sector, they are universal and compulsory and their benefits are calculated using simple formulas because of [ translation ] “the mandatory coverage of broad classes of persons and situations [that] requires that . . . operations be simplified” (Issalys and Lemieux, at p. 819; see also D. Lluelles, Précis des assurances terrestres (5th ed. 2009), at pp. 9‑10; J.‑G. Bergeron, Les contrats d’assurance (terrestre) (1989), t. 1, at p. 52). As well, they almost always include an internal administrative remedy (generally known as a “proceeding for review”) against decisions granting or denying indemnities or benefits: Issalys and Lemieux, at p. 903.

[ 35 ] It is therefore primarily or even exclusively public law that governs both the characterization of social insurance programs and the determination of the appropriate remedy for a dissatisfied person who wishes to contest a decision made in the context of such a program.

[ 36 ] The ASRA Program differs from a social insurance program. Unlike most such programs, it applies to only one sector and is neither universal nor compulsory, and its benefits are not calculated using simple formulas applicable to broad classes of persons and situations. In addition, disputes arising out of the application of the ASRA Program are not submitted to an administrative tribunal for adjudication.

[ 37 ] The ASRA Program also includes several contract‑style clauses. Section 19 provides that [ translation ] “[f]ailure to meet the conditions for eligibility throughout the participation period shall result in the resolution of the participant’s contract for the current year.” Section 101 of the program states that participants can also be excluded, in which case they must pay a penalty as provided for in s. 103:

[ translation ]

101. La Financière agricole shall exclude a participant from the Program in respect of an insurable product if the participant:

1° refuses to pay any contribution that is due;

2° refuses to submit to an inventory, an area measurement or a sampling or count of a stored or marketed crop; or

3° applies for an exclusion in writing.

[ 38 ] Regarding the penalties that result from exclusion, the first paragraph of s. 103 of the ASRA Program provides that, [ translation ] “[w]here a farming business has been excluded, La Financière agricole shall keep every amount received as a contribution in respect of the product for which the exclusion has been applied.” The second paragraph of s. 103 adds that “[a] farming business that has been excluded shall pay contract resolution fees corresponding to 25% of the contribution it paid for the last year in which it complied with the insurable minimum.”

[ 39 ] Finally, any amendments to the program that are not related to the contribution rate may not come into force until the year after they are made (s. 21 ASRA Program). Participants can therefore decide to opt out of the program for the following year if amendments proposed by La Financière are not acceptable to them.

[ 40 ] These mechanisms for terminating the contract for predetermined reasons, which create acquired rights for the current year, fall under contract law much more than public law. Moreover, Geoffroy J. of the Superior Court reached a similar conclusion in Trépanier v. Financière agricole du Québec , 2011 QCCS 1802 , a case that closely resembled the case at bar. Geoffroy J. commented as follows in analyzing La Financière’s powers under the ASRA Program:

[ translation ] Although La Financière agricole must report to the government on its management (s. 43), it can adapt its programs (ss. 20 and 22) without having to seek any form of prior administrative authorization. One reason for this autonomy is that the government and the Union des producteurs agricoles (UPA) are represented on the agency’s board of directors (s. 6 of the Act).

However, the defendant’s power is not absolute, and its programs must include the terms and conditions needed to ensure sound management of the funds it is responsible for administering. It is essential that the defendant’s clients be made aware of the contractual conditions that are binding on them as participants in the [ASRA] [P]rogram , and that is what was done in this case. [Emphasis added; paras. 59‑60 ().]

[ 41 ] In our view, the characteristics of the ASRA Program referred to by Geoffroy J. — in particular La Financière’s considerable management autonomy, limited as it is by the need to comply with the “contractual conditions” that bind La Financière and the participants — support the proposition that this program is by nature a contractual mechanism rather than a social insurance program governed by public law.

[ 42 ] These same characteristics also distinguish the ASRA Program from simple agricultural subsidy programs like the one considered by this Court in Jacobs v. Agricultural Stabilization Board , 1982 157 (SCC) , [1982] 1 S.C.R. 125. In that case, farm producers contested, inter alia , the imposition by the Agricultural Stabilization Board of ceilings on the amounts that could be granted as subsidies under the Agricultural Stabilization Act , R.S.C. 1970, c. A‑9. Laskin C.J. applied the public law rules of procedural fairness to the Board’s decisions establishing the conditions for granting subsidies, relying in particular on the fact that there “is no entitlement to subsidy until a scheme is propounded by the Governor in Council and until action is taken under the scheme by inviting the presentation of claims for subsidy” ( Jacobs , at p. 137). Laskin C.J. added that subsidies were granted without any requirement of consideration, on an “ ex gratia ” basis (p. 138).

[ 43 ] Given the contributions required from participants, the contract resolution clauses and the existence of acquired rights during an insurance year, the ASRA Program clearly differs from one under which an “ ex gratia ” agricultural subsidy is paid without consideration. Indeed, it was these very characteristics that led the Ontario High Court of Justice to find Ontario’s statutory crop insurance scheme to be a contract (of insurance) and not simply an agricultural support program: George A. Demeyere Tobacco Farms Ltd. v. Continental Insurance Co. (1984), 1984 1823 (ON SC) , 46 O.R. (2d) 423. The crop insurance scheme provided by La Financière is also governed by contract law: Brissette v. Financière agricole , 2006 QCCS 1620 . As we will explain below, there are clearly fundamental differences between a crop insurance scheme and the ASRA Program, and these differences support the proposition that the ASRA Program is not a contract of insurance within the meaning of the Civil Code of Québec . Nevertheless, it would be rather strange if one insurance program offered by La Financière were governed by private law while another were subject to a fundamentally different set of rules, those of public law.

[ 44 ] The ASRA Program can also be distinguished from certain other schemes that are subject to public law rules. One example can be seen in Canada (Attorney General) v. Mavi , 2011 SCC 30 , [2011] 2 S.C.R. 504, in which the issue was the legal status of an undertaking made by a person sponsoring a relative for immigration purposes to support the sponsored relative and to reimburse the government for any amount received by the latter as social assistance. In that case, the sponsor’s undertaking was required by statute, and any person wishing to sponsor a relative had to undertake to support the relative (para. 48). This was one reason why this Court found that the undertaking made by sponsors to the government was subject to public law, not to private law (para. 49). In the instant case, in contrast, participation in the ASRA Program is voluntary. As well, in Mavi , the undertaking signed by sponsors was incidental to the public law scheme established by the Immigration and Refugee Protection Act , S.C. 2001, c. 27, and had no purpose other than enabling relatives to immigrate as members of the family class under that public law scheme. The ASRA Program is not incidental to any statutory public law scheme.

[ 45 ] Thus, the ASRA Program is not a public law scheme but a contract. In reaching this conclusion, we wish to be clear that our reasons relate only to the scheme under consideration in this case. The determination of whether a scheme falls primarily under public law or under private law is a contextual exercise from which no extrapolation is possible. In other circumstances, it will, for example, be possible to hold, as in Mavi , that a contract entered into under a governmental scheme can be subject primarily, or even exclusively, to public law.

[ 46 ] We therefore agree with the Court of Appeal (at para. 71) that the ASRA Program has the characteristics of an [ translation ] “administrative contract”, that is, a contract to which a public authority is a party: P. Garant, with P. Garant and J. Garant, Droit administratif (6th ed. 2010), at p. 349. However, unlike the Court of Appeal, which found that the ASRA Program falls within both areas of law, that is, public law and private law (paras. 54 and 57), we are of the view that the rules that apply to this program are those of private law. Once it has been determined that the ASRA Program is a contract, we fail to see how it could be subject, even in part, to judicial review on the basis of administrative law principles or to other public law principles. All the rules needed to guide the actions of the parties under the ASRA Program can be found in private law. The courts have often applied this principle in the context of commercial relations between the government and a private party, even where the content of the contract is dictated largely by statute or regulation, which is not the case here: see, for example, Glykis v. Hydro‑Québec , 2004 SCC 60 , [2004] 3 S.C.R. 285, at paras. 18 and 30 . In our view, the courts have been right in the past to apply the rules of contract law to the ASRA Program, including those relating to defects of consent and to the scope of the obligations of the program’s participants: see, for example, Financière agricole du Québec v. Forand , 2009 QCCQ 10263 .

[ 47 ] This being said, administrative contracts are still distinguishable from contracts between private parties, since parity between the parties does not always exist. As Professor Garant explains:

[ translation ] In our law, the essential difference between a private sector contract and an administrative contract is not one of basic legal classification. Both types of contract are in fact governed by the rules of the Civil Code of Québec . . . . However, what the public interest purpose means for one of them, in contrast with the other, is that it may be subject to rules that depart from the general law, which will vary in light of the requirements of the general interest and may differ from one service to another. It is above all in this sense that an administrative contract involves privileges and constraints in comparison with a private contract; the influence of the public service is always felt; there is never absolute parity between the parties, since one represents the general interest of the community and the other represents a private interest. . . .

. . . It is thus clear that the public interest is of great importance in the government’s contracting process. This principle must guide the formation and performance of the contract and also serve as a rule of interpretation. [Footnote omitted; pp. 349‑51.]

[ 48 ] Thus, the government’s contractual relations are governed by a specific set of rules, and the public interest must be considered in interpreting such relations. When interpreting the scope of the government’s powers as a party to a contract in order to determine, for example, whether the language of the contract confers a discretion on the government, the principle of the public interest may weigh in favour of a broader discretion in implementing the scheme. This will be especially true in cases in which the contractual scheme in question has a social objective. These are not principles of public law, but considerations related to the object of the contract that may influence the interpretation of the scope of the contractual powers of the public authority in question. Along the same lines, the Court has recognized in the public tendering context that, although the tender process is in the contractual realm, the government has broad powers that enable it to “include stipulations and restrictions and to reserve privileges to itself”: Martel Building Ltd. v. Canada , 2000 SCC 60 , [2000] 2 S.C.R. 860, at para. 89 , quoting Colautti Brothers Marble Tile & Carpet (1985) Inc. v. Windsor (City) (1996), 36 M.P.L.R. (2d) 258 (Ont. Ct. (Gen. Div.)), at para. 6. There are of course distinctions to be made between the ASRA Program and a call for tenders. We will not comment on the role of public law in the regulation of the tendering process. Nevertheless, the contract law principle identified in Martel Building applies in the case at bar. In interpreting the scope of the powers conferred on La Financière by its contract with ASRA Program participants, the public interest and La Financière’s social purpose are considerations that must be taken into account.

[ 49 ] The government’s discretion nonetheless has its limits. In the context of an administrative contract like the one at issue in this case, those limits do not derive from the public law duty of procedural fairness but are instead based on good faith and contractual fairness, as is explained, once again, by Professor Garant:

[ translation ] . . . awarding or amending a contract is a purely administrative matter that does not require the application of any of the principles of natural justice, procedural fairness and legitimate expectations. The duty of good faith will of course be sanctioned by the courts, as will the government’s duty to treat other contracting parties fairly. But what is in issue here is contractual fairness, not procedural fairness in the public law sense. [Footnote omitted; p. 354.]

[ 50 ] The Court applied similar principles in the tendering context in Martel Building , explaining that such contracts placed the government under an implied obligation to “treat all bidders fairly and equally” (para. 88). This principle can be readily adapted to the context of this appeal. La Financière must treat the other contracting parties fairly and exercise its discretion in good faith. The source of the contractual obligation to treat ASRA Program participants fairly is the same as in Martel Building , namely “the presumed intentions of the parties” (para. 88). It must be presumed that the parties to the ASRA Program intended to include in it an obligation of contractual fairness, which “is consistent with the goal of protecting and promoting the integrity of the [legal] process [at issue], and benefits all participants” (para. 88). In Quebec law, this amounts to an application of the rule provided for in art. 1434 C.C.Q. that “[a] contract validly formed binds the parties who have entered into it not only as to what they have expressed in it but also as to what is incident to it . . . in conformity with . . . equity”. As for the obligation to act in good faith, it flows from arts. 6, 7 and 1375 C.C.Q.

[ 51 ] In sum, like the contract itself, the fairness of the exercise of that discretion must be assessed in light of the public interest and La Financière’s social purpose. Contract law will nonetheless apply to make La Financière accountable for the financial impact of its decisions if they fail to satisfy the requirements of good faith and contractual fairness. This is the legal framework that must guide the analysis of La Financière’s actions in this case. Before analyzing those actions having regard to that contractual legal framework, we must first determine whether the rules specific to the contract of insurance apply to it.

(2) Is the ASRA Program Subject to the Rules Applicable to the Contract of Insurance?

[ 52 ] The appellants argue that the ASRA Program is a contract of insurance within the meaning of art. 2389 C.C.Q. They submit that farm insurance is based on the principles of mutuality and risk applicable to the contract of insurance and can therefore be distinguished from other financial assistance programs, such as those that involve the granting of subsidies. This means that the rule of interpretation based on the reasonable expectations of the insured must be applied in interpreting s. 88(3) of the ASRA Program.

[ 53 ] La Financière counters this position by arguing that what the ASRA Program covers is not a risk, but a contingency, given that there are no uncertain events for participants, but a recurring situation that leads to the application of the program on a regular or even permanent basis for certain products. La Financière also notes that participants cannot receive a personalized indemnity, since the amounts to which they are entitled are calculated not on the basis of losses incurred by their businesses, but on the basis of the model of a benchmark farm.

[ 54 ] In this Court, the parties’ arguments on this point focused on a comparison between the ASRA Program and the Programme d’assurance récolte also offered by La Financière. The Programme d’assurance récolte , like equivalent crop insurance programs in the other provinces, has been treated by the courts as a contract of insurance that is subject to all the rules that apply to such contracts: Brissette ; Demeyere ; Rollo Bay Holdings Ltd. v. Prince Edward Island Agricultural Development Corp. (1993), 1993 1777 (PE SCAD) , 110 D.L.R. (4th) 132 (P.E.I.S.C. (App. Div.)). The respondent argues that the ASRA Program is sufficiently different from a crop insurance program to warrant different legal treatment, whereas the appellants argue that the same rules of interpretation apply to both programs.

[ 55 ] In our view, the ASRA Program is not a contract of insurance and cannot be subject to the rules applicable to such contracts. It does not have the three main characteristics of a contract of insurance set out in art. 2389 C.C.Q. , namely (i) an obligation on the client to pay a premium or assessment; (ii) the occurrence of a risk; and (iii) an obligation on the insurer to make a payment to the client if the insured risk occurs. The reasons for this conclusion are as follows.

[ 56 ] First, it is agreed that ASRA Program participants have an obligation to pay a premium or assessment. This is not changed by the fact that the government provides most of the funding for the program through La Financière. That being said, it may happen, where the “risk” (net annual income of the benchmark farm less than the stabilized net annual income) does not occur for a certain period and the amounts paid into the fund are sufficient to ensure its sustainability, that no contributions are made in respect of a designated product. For example, no contribution was required from ASRA Program participants in 2003 in respect of “potatoes”. Moreover, the program provides that any surplus must be redistributed among participants in proportion to their contributions (s. 13). This means that one of the characteristics referred to in art. 2389 C.C.Q. is lacking here, since, according to the principle of mutuality, the contract of insurance concept presupposes that the client continues to be a debtor in respect of the premium even if the risk does not occur: Lluelles, at p. 25.

[ 57 ] Second, the implementation of the ASRA Program does not depend on the occurrence of a risk within the meaning of art. 2389 C.C.Q. What is described here as a risk is actually based on the net annual income of the benchmark farm, which La Financière calculates itself in exercising its broad discretion. La Financière therefore exerts control over the occurrence of the “risk” when it adjusts and fixes the net annual income on the basis of statistical studies or of [ translation ] “other data it deems relevant” (s. 87 para. 3 ASRA Program). In this sense, the occurrence of the “risk” under the program depends on the will of La Financière. Yet it is well established that the occurrence of the risk under a contract of insurance must be [ translation ] “an event that is uncertain and does not depend exclusively on the will of the parties”: Lluelles, at p. 186.

[ 58 ] Lastly, it is almost certain that the stabilized net annual income will ultimately be higher than the net annual income. The evidence shows that, over a 30‑year period — from 1979 to 2009 — La Financière paid compensation every year for certain products, one example being “feeder cattle”. Moreover, s. 92 of the ASRA Program provides that La Financière may make [ translation ] “advance payments on the compensation to be paid” on the basis of “forecasts” of the amounts that will be paid in the future. It can be seen from the cases that La Financière avails itself of that power: Forand , at paras. 12‑13 (); Financière agricole du Québec v. Coddington , 2013 QCCQ 6238 , at paras. 5‑6 (). This means that the “risk” is foreseeable not only as regards its occurrence, but also often as regards the time of its occurrence. As Professor Lluelles states, [ translation ] “if the promised payment is based not on the occurrence of a risk but on the occurrence of an event that is certain in terms of its occurrence, its prematurity or its severity, the contract in question is not one of insurance” (p. 25 (footnote omitted)).

[ 59 ] Furthermore, several provisions of the Civil Code of Québec relating to the contract of insurance are foreign to the essential elements of the ASRA Program. For example, there is no question of the insured giving notice of the “loss”, which first becomes known to La Financière when it determines that the stabilized net annual income is higher than the net annual income of the benchmark farm for a covered product. The obligations under arts. 2470 to 2474 C.C.Q. concerning notice of the “loss” by the insured cannot therefore apply to the ASRA Program. Nor is it possible to speak of a “material change in the risk” caused by the insured within the meaning of arts. 2466 to 2468 C.C.Q. or of “initial representations” of the insured in the context of arts. 2408 to 2413 C.C.Q. These are concepts that are central to the regulation of insurance.

[ 60 ] In this regard, the ASRA Program differs from the Programme d’assurance récolte , which accords very well with several provisions applicable to the contract of insurance. The crop insurance program protects participants from [ translation ] “uncontrollable” climatic events (s. 27). Under that program, indemnities are calculated on an individual basis (s. 38), unlike the compensation provided for in the ASRA Program, which is instead calculated on the basis of a hypothetical benchmark farm. The crop insurance program requires the existence of an individual loss of which the insured must give notice in accordance with arts. 2470 to 2474 C.C.Q. (ss. 35 to 37). Sections 10 and 15 of the Programme d’assurance récolte refer to the initial representations of the insured, while s. 42 refers to the concept of a material change in the risk caused by the insured. Whereas the Programme d’assurance récolte is compatible with arts. 2408 to 2413 and 2466 to 2468 C.C.Q. , the ASRA Program is not.

[ 61 ] It is true that the terminology of contracts of insurance is sometimes used in the ASRA Program. For example, s. 78 refers to an [ translation ] “insurable product”, para. 2 of s. 21 to an “insurance year” and s. 22 to an “insurance certificate”. In our view, the use of that terminology is not conclusive, however. When properly analyzed, the ASRA Program cannot be characterized as a contract of insurance. It is therefore not subject to the rules specific to such contracts.

[ 62 ] Since the ASRA Program is not a contract of insurance, it follows that the rule of interpretation based on the reasonable expectations of the insured does not apply. We nonetheless consider it appropriate to clarify the extent to which that rule applies in Quebec law, given that the parties to this appeal and the companion appeal, Lafortune v. Financière agricole du Québec , 2016 SCC 35 , [2016] 1 S.C.R. 1091, present different views and refer to a debate about the rule among the commentators and the courts. In both appeals, ASRA Program participants argue that the reasonable expectations rule applies in its [ translation ] “maximum dimension” and that La Financière’s powers are limited by their expectations regardless of whether the contract is ambiguous. La Financière counters that the rule must apply only in its “minimum dimension”, that is, only where there is ambiguity.

[ 63 ] The scope of this rule, which originates in U.S. insurance law, was explained by Cory J., dissenting, but not on this point, in Brissette Estate v. Westbury Life Insurance Co. , 1992 32 (SCC) , [1992] 3 S.C.R. 87, at p. 102. He mentioned that this rule of interpretation was applied in the United States in three ways: (1) to resolve any ambiguity in the terms of the contract in favour of the insured in order to satisfy his or her reasonable expectation; (2) to give the insured a right to all the coverage he or she was entitled to expect, unless there was an “unequivocal plain and clear manifestation of the company’s intent to exclude coverage”; or (3) to give the insured such coverage even in cases in which “painstaking study of the policy provisions would have negated those expectations” (p. 103). The first and third of these scenarios correspond, respectively, to what some authors have called the “minimum” and “maximum” dimensions of the doctrine: D. Lluelles and B. Moore, Droit des obligations (2nd ed. 2012), at pp. 922‑23; M.‑C. Thouin, “La théorie de l’attente raisonnable de l’assuré” (1997), 64 Assurances 545, at p. 551. However, none of them allows the meaning of a clear provision to be disregarded in favour of the expectations of the insured, except, in the third case, insofar as the interpretation of the provision requires “painstaking study” to determine its true meaning.

[ 64 ] The Court has accepted the first formulation (the minimum dimension) of the reasonable expectations rule in a number of cases, holding that, in Canada, the rule applies only in the event of ambiguity: Jesuit Fathers of Upper Canada v. Guardian Insurance Co. of Canada , 2006 SCC 21 , [2006] 1 S.C.R. 744, at paras. 27 and 29 ; Reid Crowther & Partners Ltd. v. Simcoe & Erie General Insurance Co. , 1993 150 (SCC) , [1993] 1 S.C.R. 252, at p. 269; National Bank of Greece (Canada) v. Katsikonouris , 1990 92 (SCC) , [1990] 2 S.C.R. 1029, at p. 1043. The Quebec Court of Appeal has consistently reached the same conclusion: Excellence (L’), compagnie d’assurance‑vie v. Desjardins , 2005 QCCA 1035 , [2005] R.R.A. 1085, at para. 11 ; Affiliated FM Insurance Co. v. Hafner Inc. , 2006 QCCA 465 , at para. 47 (); Souscripteurs du Lloyd’s v. Alimentation Denis & Mario Guillemette inc. , 2012 QCCA 1376 , at para. 38 ().

[ 65 ] We are of the view that in Quebec law, the reasonable expectations rule must apply solely in its minimum dimension, that is, only where there is ambiguity. The Civil Code of Québec contains a series of rules for resolving difficulties in the interpretation of contracts. The reasonable expectations rule can certainly be added to those rules of interpretation for contracts of insurance, but only for the interpretation of an ambiguous clause and for the purpose of identifying the common intention of the parties (art. 1425 C.C.Q. ).

[ 66 ] This being said, in the case at bar, because the ASRA Program is an innominate administrative contract that does not have the characteristics of a contract of insurance, the rules of contractual interpretation set out in arts. 1425 to 1432 C.C.Q. must be applied to determine the outcome of the appeal. We must refer to those rules to resolve the central issue, that is, whether it was open to La Financière to collectively link the amounts received under the two federal assistance programs in respect of which the appellants brought their action.

B. Was it Open to La Financière to Collectively Link the Amounts Received From the Federal Government Under the COPB and Kickstart Programs?

(1) Interpretation of Section 88(3) of the ASRA Program

[ 67 ] Section 88(3) of the ASRA Program does not specify how compensation received under government assistance programs is to be linked. It states only that, for this purpose, La Financière must consider [ translation ] “[a]ny amounts to which a participant is entitled” under such programs.

[ 68 ] The appellants argue that the words “a participant” [ translation ] “refe[r] to each participant considered individually and not . . . to a hypothetical participant on the benchmark farm” (A.F., at para. 63). They add that s. 88(3) must be read in conjunction with s. 2 of the ASRA Program, in which “participant” ( adhérent ) is defined as [ translation ] “a farming business . . . that participates in the Program”. In the appellants’ opinion, s. 88(3) therefore requires individual linkage. Our colleague agrees with them. In her view, the wording of s. 88(3) means that La Financière cannot take account of amounts to which the participant is not entitled.

[ 69 ] La Financière replies that s. 88(3) must be interpreted in light of the program as a whole, which is based on the average benchmark farm, and that collective linkage should therefore be the rule.

[ 70 ] To begin, we cannot agree with the appellants’ main argument that the words [ translation ] “amounts to which a participant is entitled” mean that La Financière must always link the amounts individually. It is common ground that La Financière has used both collective and individual linkage in the past. In reality, as will be seen below, individual linkage has been the exception rather than the rule. Each time La Financière has linked amounts collectively, it has inevitably penalized participants for amounts to which they were not entitled, as it has in such cases attributed to all participants amounts that were calculated on the basis of what the benchmark farm would have received. An interpretation requiring that amounts be linked individually in every case would conflict with that past practice and would as a result call into question most of the decisions made by La Financière over the past 15 years. It would also put s. 88(3) in conflict with the whole of the ASRA Program, which is based on the collective concept of a benchmark farm. In our opinion, s. 88(3) cannot be interpreted so literally.

[ 71 ] Nor, from this point of view, is the concept of double compensation to which our colleague refers (at paras. 115‑21) a conclusive interpretive guide for the resolution of this appeal. If the purpose of s. 88(3) of the ASRA Program were to avoid double compensation solely from the perspective of each participant, amounts would have to be linked individually in every case, as the linked amounts would always have to be identical to what each participant received. The evidence shows that this is quite simply not the case. The fact that La Financière has most often linked amounts collectively shows that its purpose has not been to avoid double compensation solely from the perspective of each participant considered individually. The purpose of s. 88(3) is also to account for other income that could have an impact on the financial needs of the average benchmark farm. That is one purpose that La Financière has pursued, as it was free to do.

[ 72 ] In our opinion, to determine what meaning should be attributed to this provision of the ASRA Program, it is instead necessary to consider the rules of interpretation set out in the Civil Code of Québec . The application of those rules leads to the conclusion that the contract does not require that amounts be linked individually in cases such as this one but, on the contrary, gives La Financière the discretion to decide what linkage method to employ.

[ 73 ] Two rules of interpretation are particularly relevant to the identification of the common intention of the parties to the ASRA Program. One of them, set out in art. 1427 C.C.Q. , is that each clause of a contract must be interpreted in light of the others so that each is given the meaning that flows from the contract as a whole. To interpret s. 88(3) literally would be to disregard this first rule. The other rule, in art. 1426 C.C.Q. , is that, when a court interprets a contract, the interpretation which has already been given to it by the parties or which it may have received, and usage in such matters, are all to be taken into account. Favouring individual linkage would be contrary to this second rule.

[ 74 ] Although a reading of s. 88(3) of the ASRA Program in the context of the program as a whole does not eliminate the possibility of linking amounts individually, it does lead to the conclusion that collective linkage must be preferred, as Savard J.A. rightly observed (para. 83). Section 88(3) is in Division XI, entitled [ translation ] “Farm Models”, which is a collective concept. Moreover, ss. 86 and 92 of the ASRA Program clearly state that the net annual income used to calculate the compensation, which includes amounts received from government sources, is that of a “specialized benchmark farm for each of the products or classes of products”.

[ 75 ] The general structure of the program and the contract as a whole thus support the conclusion that collective linkage is normally required. However, given that La Financière has sometimes linked amounts individually in the past, it will be helpful to consider that past practice in order to determine its scope and any influence it has had in terms of usage. This analysis reveals the limits placed on individual linkage by La Financière.

[ 76 ] On this point, the trial judge found that La Financière had in the past linked amounts individually where, as in this case, government assistance had been paid directly to producers. She found that La Financière had linked compensation collectively only where a government agency had paid a single sum to La Financière’s ASRA Program fund (paras. 133‑34). While it is true that the trial judge was relying on these facts to justify her application of the rule based on the reasonable expectations of the insured, which we are excluding here, past practice remains relevant to the interpretation of the content of the contract because of art. 1426 C.C.Q.

[ 77 ] With respect, we find that the trial judge made a palpable and overriding error in concluding that in light of past practice, there was a clear correlation between direct payments and individual linkage. The evidence does not establish that amounts paid directly to producers have always been linked individually. Furthermore, the evidence shows that several other considerations have affected the choice of linkage method in the past, the most important being the number of ASRA Program participants that received the amounts to be linked.

[ 78 ] In support of her conclusion that amounts had been linked individually [ translation ] “in the case of direct payments to producers”, the trial judge listed five programs with the acronyms TISP, FIPP, BSE‑7, GOPP and CAIS (para. 134). She referred in this regard to a table prepared by La Financière that showed the past programs and the linkage method used for each of them. It can be seen from the table that FIPP and BSE‑7 were a single program, which means that there have only been four programs in the past in respect of which amounts were linked individually. The table also indicates that, for two of those four programs (TISP/BSE‑4 and FIPP/BSE‑7), the linkage of amounts was mixed: collective for some products and individual for others. This was confirmed by the testimony of two representatives of La Financière, André Houle and Alain Pouliot. Finally, these two representatives also testified that amounts paid under the CSRA, a provincial program shown in the table prepared by La Financière, which were linked collectively, had been paid in the form of direct payments to producers in accounts opened for individual participants.

[ 79 ] This evidence shows that there was no systematic correlation between direct payments to producers and individual linkage. La Financière has already had recourse to collective linkage for several amounts that had been granted directly to producers. It is clear that this has sometimes had the effect of attributing to certain producers income that they had not received.

[ 80 ] La Financière adds that there are other considerations that explain decisions made in the past to link amounts individually. The main one is that amounts from programs that benefited a small number of participants were linked individually. The evidence thus shows that it was on this basis that La Financière individually linked amounts received under GOPP and CAIS, the only two programs for which amounts were linked on a strictly individual basis. The evidence also shows that the number of participants that received money was a decisive factor in the choice of individual linkage for certain products in the case of TISP/BSE‑4 and FIPP/BSE‑7. This was at the very least the case for “Lambs” under FIPP/BSE‑7. Mr. Houle’s testimony suggests that it was also the case more generally under those two programs.

[ 81 ] It thus appears that the decision to link amounts individually was usually based on the number of participants that received the amounts in question. If by chance only a few participants received a particular government benefit, it would have been unfair to link the amounts collectively, as notional income amounts would have been attributed to many participants who had in fact received nothing at all. In such cases, it was a matter not of there being a difference between the notional amounts and amounts that had actually been received, but of income being attributed where none had even been received.

[ 82 ] The evidence therefore establishes that there has never been any correlation in the past between direct payments to producers and individual linkage. It follows that past practice has not been consistent enough to support a conclusion that individual linkage was the appropriate method in this case.

[ 83 ] We are therefore of the view that La Financière was under no statutory or contractual obligation to link the amounts individually in this case. Under s. 88(3), La Financière had a discretion in respect of the choice of linkage method. Neither the contract as a whole nor past practice supports a conclusion that one or the other of the two linkage methods had to be employed. In the absence of a provision circumscribing La Financière’s powers, s. 88(3), as it read at the relevant time, must be found to have given La Financière the choice of linking the amounts collectively or individually having regard to what was fair in the circumstances and to La Financière’s mission. In s. 88(3), La Financière has retained the broad discretion it has under ss. 19 and 22 AFAQ . The amendments made to s. 88(3) in November 2008 are not conclusive in this regard. The clarification made in those amendments, to the effect that La Financière will link amounts collectively unless it deems it appropriate to choose another method, corresponds precisely to the proper interpretation of the previous version of s. 88(3), which simply read [ translation ] “amounts to which a participant is entitled”. The amendments did not change the terms of the contract; they simply confirmed the meaning of the former version.

[ 84 ] Given this conclusion, there is no need to apply the contra proferentem rule of interpretation, which requires that any difficulties in interpreting a contract of adhesion be resolved in favour of the adhering party (art. 1432 C.C.Q. ). That rule is an interpretive mechanism of last resort. In Consolidated‑Bathurst Export Ltd. v. Mutual Boiler and Machinery Insurance Co. , 1979 10 (SCC) , [1980] 1 S.C.R. 888, at pp. 899‑900, the Court confirmed that, in Quebec law, the other rules of contractual interpretation (like the rule that each clause of a contract is interpreted in light of the others so that each is given the meaning derived from the act as a whole) take precedence over that of contra proferentem (see also Industrielle, Compagnie d’Assurance sur la Vie v. Bolduc , 1978 171 (SCC) , [1979] 1 S.C.R. 481, at p. 493).

[ 85 ] Moreover, the contra proferentem rule would be difficult to apply in the instant case, given that the ASRA Program covers 16,747 participants, only some of whom, like the 137 appellants, were allegedly disadvantaged by La Financière’s decision to link the amounts collectively. It would therefore be quite incongruous to seek to interpret the contract in favour of the adhering party given that the many participants in the program have widely varying and sometimes also conflicting interests.

(2) La Financière’s Exercise of Its Discretion

[ 86 ] That being said, although La Financière had the discretion to link the amounts collectively, it was required to exercise that discretion in accordance with the requirements of good faith and contractual fairness. In our view, the Court of Appeal was right to conclude that La Financière did so here.

[ 87 ] The decision to link the amounts collectively was made following extensive consultations with representatives of farm producers. Thus, on June 6, 2007, representatives of La Financière met with members of the Table sur la sécurité du revenu of the Union des producteurs agricoles (“UPA”) to discuss the COPB and Kickstart programs and the appropriate linkage method. We note that the Table sur la sécurité du revenu, an internal UPA committee made up of all presidents of specialized federations whose products are covered by the ASRA Program and of several other presidents of specialized and regional federations, is the body that La Financière generally approaches in order to hold discussions with representatives of farm producers. At the end of that meeting on June 6, 2007, everyone was in agreement that amounts received under the COPB and Kickstart programs would be linked collectively on the basis of the characteristics of the specialized benchmark farm. La Financière’s board of directors, which was chaired by the UPA’s president and 5 of the 11 members of which represented that organization, subsequently approved that decision to link the amounts collectively.

[ 88 ] Furthermore, La Financière had conducted impact simulation studies that showed that most of the program’s participants would benefit from its decision to link the amounts collectively. It did not save money in linking them collectively. According to the evidence in the record, linking the amounts collectively cost some $1,620,472 more in this case than linking them individually would have done.

[ 89 ] The evidence also shows that, in the past, individual linkage has been chosen when a small number of ASRA Program participants received the amounts to be linked. In this case, however, this factor favoured collective linkage, given that all ASRA Program participants were eligible to receive amounts under the two federal programs, COPB and Kickstart.

[ 90 ] In addition, the evidence shows that another consideration played a major role in La Financière’s decision. In this case, unlike with the vast majority of past programs, benefits under the COPB and Kickstart programs were calculated on the basis of overall sales, which included sales of products that were not covered by the ASRA Program. By linking those amounts collectively, La Financière could therefore break down the amounts received by the benchmark farm by product and exclude any amounts allocated for products that were not covered by the ASRA Program. It thus avoided penalizing all the participants in that it did not take into account income generated by products that were not covered but were still part of their agricultural operations. Less compensation received under the ASRA Program was deducted if only income from the 10 designated products or classes of products was taken into account. The evidence shows that such an exercise would have been impossible had the amounts been linked individually, because La Financière did not have detailed information about each participant’s products for which amounts had been granted under the COPB and Kickstart programs, whereas that information was available in respect of the benchmark farm.

[ 91 ] La Financière’s choice to link the amounts collectively was quite compatible with the nature of the ASRA Program, which is structured around the concept of the average benchmark farm. By linking them collectively, La Financière favoured the smallest producers, whose allowable net sales were, like those of the benchmark farm, below the caps imposed under the two federal programs. This situation was consistent with La Financière’s mission, namely to “support and encourage the development of the agricultural and agro‑food sector within the perspective of sustainable development” ( s. 3 AFAQ ).

[ 92 ] It is true that La Financière’s decision to link the amounts collectively had an adverse effect on the appellants, as the trial judge observed with supporting figures (paras. 68‑80). For example, the appellant Ferme Vi‑Ber inc. received $12,240 and $51,391.30 from the federal government under the COPB and Kickstart programs. Because the amounts were linked collectively, the company was attributed income of $47,363.34 and $96,300.65, respectively, under those two programs. As with the other 136 appellants, these were substantial differences as a result of which Ferme Vi‑Ber inc. received less compensation than it would have received had the amounts been linked individually. However, this is not enough to support a finding that the decision to link the amounts collectively was unfair or that La Financière did not act in good faith. The ASRA Program is an ongoing program. Producers that are disadvantaged today could very well be advantaged by other decisions in the future. For example, the appellants could be advantaged by a situation in which amounts are linked individually, such as the one alleged to be unfair in Trépanier , which Geoffroy J. of the Superior Court refused to find to be invalid (paras. 65-66).

[ 93 ] La Financière acted in good faith and took the necessary steps to weigh the impact on participants of the decision to link the amounts collectively in order to ensure that the decision was fair. It is clear that the specific features of each program and the impact on all the participants help determine which linkage method is appropriate. That impact is dependent on a host of variables, including year‑to‑year fluctuations in allowable net sales, the timing of the establishment of a farming business, an outbreak of disease at a particular time and the production sector that is involved. In the case at bar, La Financière also states that it reduced the contribution rate of hog and piglet producers to minimize the impact of the chosen linkage method for that sector. Indeed, it asked the trial judge to consider this in any award of damages (trial judge’s reasons, at paras. 105‑6). This clearly shows why, where there is a reasonable explanation for the choice that was made and where there is no evidence of bad faith in the entity’s actions, the courts should avoid interfering too hastily in such complex matters.

[ 94 ] We conclude that La Financière exercised its contractual powers in accordance with the requirements of good faith and contractual fairness. That being so, it was open to La Financière in fixing the compensation payable to the appellants to choose to link the amounts they had received under the two federal financial assistance programs collectively.

VI. Disposition

[ 95 ] We would therefore dismiss the appeal. Like the Court of Appeal, we would not award costs in light of the relationship between the parties.

English version of the reasons delivered by

[ 96 ] Côté J. (dissenting in part) — Although I agree with my colleagues on the nature of the Programme d’assurance stabilisation des revenus agricoles , (2001) 133 G.O. 1, 1336 (“ASRA Program”), and also agree that it is not necessary to apply public law principles to decide the case, I am of the view that the only determinative issue in this case is one of contractual interpretation.

[ 97 ] The reason why producers participate in the ASRA Program is simple: they expect no more and no less than to receive the full compensation they are owed in return for paying their contributions. When La Financière agricole du Québec (“La Financière”) deprives them of all or part of the compensation to which they are entitled, contrary to the express provisions of the contract it has itself imposed on them , judicial intervention is warranted.

[ 98 ] The common intention of the parties, the overall scheme of the ASRA Program and past practices confirm that La Financière contravened the program in deciding to subtract excessively high amounts of notional income from the amounts to which participants were otherwise entitled. It did this not to ensure that participants did not receive double compensation for a single loss, but to redistribute the amounts that were received. The amounts attributed to participants bore no relation to the amounts to which they were entitled under federal programs. La Financière itself admits it knew that, in employing the calculation method it did, it was disregarding the caps specified in those programs.

[ 99 ] Regardless of whether the contract is characterized as a contract of insurance or as an innominate contract that falls under both public law and private law, the result is the same. It is true that the appellants participated voluntarily in the ASRA Program, but insofar as that program involves a contract of adhesion imposed by La Financière, that is, a contract that is not negotiated with participants, I find that the characterization makes no difference. If there is any ambiguity, it must be resolved in favour of the adhering party in accordance with art. 1432 of the Civil Code of Québec (“ C.C.Q. ”). The rule of interpretation of reasonable expectations adds nothing to the existing rules of interpretation.

[ 100 ] In my opinion, the appeal should therefore be allowed, but only in part.

I. Background

[ 101 ] The appellants are all participants in the ASRA Program administered by La Financière. The purpose of the program is to guarantee them a positive net annual income for certain products ( s. 1 ). When the market price of a covered product is lower than production costs, a participant is entitled to compensation from the Fonds d’assurance stabilisation des revenus agricoles (“Fund”). Participants pay a third of the total contributions to the Fund, and La Financière pays two thirds (s. 80). The amount of a participant’s contribution may be reviewed annually (s. 6).

[ 102 ] The compensation to which a participant is entitled is determined on the basis of the participant’s volume of insurable products. It does not depend on the participant’s particular financial situation but is instead based on the performance of a [ translation ] “specialized benchmark farm” for each of the products. It corresponds to the difference between the benchmark farm’s net annual income and stabilized annual income multiplied by the participant’s volume of insurable products, that is, the number of units of products marketed by the participant:

[ translation ] The net annual income essentially corresponds to the earnings of the specialized benchmark farm, a description of which in relation to each insurable product is set out in Table 4 of section 86. The net annual income is determined by La Financière agricole on the basis of an economic study of a specialized benchmark farm for each product (or secondary product) covered by the Asra Program (s. 87). For each unit of an insurable product, it corresponds to annual receipts minus cash disbursements and depreciation, which are determined as specified in the Asra Program (ss. 86 to 88 and 91).

The stabilized net annual income is also determined for each unit of a product covered by the Asra Program after consultation with the representatives of farming businesses (ss. 2 and 89). Technically, it is calculated as a percentage (ranging between 70% and 90%, depending on the product) of the “average annual regular salary of a skilled worker” (or a portion of that salary ranging between 0.38 and 1.35), which was fixed at $42,461 in 2000 (and indexed thereafter). [Emphasis in original; footnote omitted.]

(Court of Appeal’s reasons, 2014 QCCA 1886 , at paras. 24‑26 ())

[ 103 ] It sometimes happens that during the year, as in this case, additional amounts from other government agencies are paid to individual participants, as opposed to a situation where a global amount is paid to La Financière. These are amounts to which participants are personally entitled. The ASRA Program expressly authorizes La Financière to take such amounts into account in its calculations to ensure that participants do not receive double compensation for a single loss (s. 88). The problem here is whether La Financière can, in performing its calculations, totally disregard the amounts that were actually received or whether it can consider notional amounts where the amounts were granted to participants individually.

[ 104 ] In 2007, the federal government established two programs to provide assistance to Canadian farming businesses: the Cost of Production Benefit (“COPB”) program and the AgriInvest Kickstart (“Kickstart”) program. At issue in this case is the linkage between those programs and the ASRA Program. Under the programs in question, amounts were paid directly to producers and were allocated to each producer individually, although La Financière was responsible for administering them in Quebec. Each program imposed a cap on allowable net sales (“ANS”) above which participants could not be compensated. The maximum amounts of ANS were set at $450,000 ($12,240 per producer) for the COPB program and $3 million ($96,000 per producer) for the Kickstart program.

[ 105 ] La Financière decided to take this direct federal assistance into account by considering the amounts the benchmark farm would theoretically have received for each product. It dealt with those amounts as if they constituted a global sum that had been paid to it and that it could redistribute as it saw fit. The federal assistance was taken into account in calculating the benchmark farm’s stabilized income in light of the characteristics of that farm, including its ANS. The resulting figure was then multiplied, for each participant, by the number of units of insurable product.

[ 106 ] The parties characterized this approach as [ translation ] “collective linkage”. Under it, the caps imposed by these programs on the insurance coverage of ASRA Program participants were completely disregarded. The evidence shows that the effect of this “linkage” method was to subtract from the compensation otherwise owed to many participants amounts higher than what they had actually received in federal benefits.

[ 107 ] In some cases, there was a huge difference between the amounts participants actually received and the amounts that were attributed to them. For example, Ferme Vi‑Ber received a COPB benefit of $12,240 but had $47,363.34 subtracted from the compensation owed to it under the ASRA Program. In the case of Alfred Couture Limitée, there was an even greater difference between the amount it actually received and the COPB amount that was attributed to it. Even though its assistance under the COPB program was capped at $12,240, La Financière attributed $94,170.38 to it. Similar differences could be seen in relation to the Kickstart program. Thus, Ferme Vi‑Ber received a Kickstart benefit of $51,391.30 in its AgriInvest account, but La Financière attributed $96,300.65 to it. The amount attributed to Inter Agro inc., $800,146.49, was completely out of proportion to the $96,000.00 it received: trial judge’s reasons, 2012 QCCS 284 , at paras. 70 and 78 ().

[ 108 ] The appellants contest the linkage method applied by La Financière and the resulting reduction in their compensation. They argue that La Financière not only disregarded s. 88(3) of the ASRA Program, but also diverted it from its purpose, which is simply to avoid double compensation, not to deprive them of compensation they are otherwise owed.

[ 109 ] In my view, the appellants are right.

II. Analysis

[ 110 ] Over time, the federal government has multiplied its initiatives designed, inter alia , to better respond to the realities of the agricultural sector. Each of these initiatives has its own characteristics and could result in double compensation if it is not properly taken into account, which is why s. 88(3) of the ASRA Program was adopted:

[ translation ]

88. The following items shall be taken into account in calculating annual receipts:

3° Any amounts to which a participant is entitled on the basis of the volume of marketed products and secondary products and that are granted by government agencies in the form of price compensation for the insurable product or under a government farm business risk management program.

[ 111 ] I agree that La Financière must, in administering the ASRA Program, be able to adjust its stabilization insurance initiatives on the basis of the various measures that are adopted by other government agencies on an ad hoc or recurring basis. However, it must do so in accordance with the provisions of the ASRA Program, including s. 88(3).

[ 112 ] As La Financière itself admits, there are limits on its ability to adjust to the initiatives of other granting organizations and to take into account the amounts to which participants are entitled under other programs. Once the ASRA Program had been adopted, La Financière had to comply with the rules of the game it had itself established. Although the limits are not numerous, they do exist. It goes without saying that La Financière must in all circumstances treat the participants fairly and act in good faith, but it must also comply with the ASRA Program’s express provisions.

[ 113 ] In the case of amounts received directly from other granting organizations that can be considered to be annual receipts, La Financière’s discretion is expressly limited. Section 88(3) of the ASRA Program provides that, in calculating annual receipts, La Financière must take into account [ translation ] “[a]ny amounts to which a participant is entitled on the basis of the volume of marketed products and secondary products”.

[ 114 ] In this case, La Financière’s conduct is problematic from two points of view: the purpose of s. 88(3) and its wording.

A. Purpose of Section 88(3): To Avoid Double Compensation

[ 115 ] In its factum and at the hearing in this Court, La Financière acknowledged that the purpose of s. 88(3) is to avoid double compensation for a single loss. Its representatives also acknowledged this in their testimony. When, as in this case, the federal government grants and pays a specific amount directly to a participant, the linkage must be based on that amount.

[ 116 ] By applying s. 88(3) to attribute excessively high notional amounts to some participants, La Financière diverted that provision from its purpose. La Financière used s. 88(3) not to avoid double compensation but to arbitrarily reduce the compensation to which many participants would otherwise have been entitled under the ASRA Program.

[ 117 ] Although s. 88(3) authorizes La Financière to ensure that participants do not receive double compensation, it does not permit La Financière to attribute notional amounts to some participants in order to overcompensate other participants for policy reasons . In other words, La Financière deprived some participants of the compensation they were owed in order to add to the compensation paid to others . Thus, not only did the chosen linkage method not make it possible to avoid double compensation, it also prevented many participants from receiving the full compensation to which they were entitled under the ASRA Program. That full compensation is the very reason why participants choose to participate in that program.

[ 118 ] That decision is particularly troubling for participants, who, since they are unable to negotiate the terms of the ASRA Program, must be able to rely on the program’s provisions to determine whether it is in their interest to participate. While it is true that La Financière’s mission is broad, that mission does not authorize it to subvert the purpose of the ASRA Program by assuming a discretion it does not have. La Financière’s mission does not authorize it to penalize its participants for participating in other programs.

[ 119 ] To maintain, as my colleagues do, that the only purpose of s. 88(3) is “to account for other income that could have an impact on the financial needs of the average benchmark farm” (para. 71) amounts to giving La Financière an absolute discretion that the words of s. 88(3) do not authorize. This line of reasoning implies that La Financière can disregard the inherent limits of federal programs in order to redistribute amounts, even notional ones.

[ 120 ] La Financière admits, at para. 14 of its factum, that the purpose of s. 88(3) is to avoid double compensation in respect of the amounts to which farming businesses that participate in the ASRA Program are entitled:

[ translation ] Given that farming businesses that participate in ASRA can also benefit from federal farm income support programs, the Respondent is required to take amounts to which a participant would be entitled under those programs into account in order to avoid double compensation . This follows from s. 88, s. 86 second para. and the definition of “annual receipts” set out in s. 2 of ASRA, which are quoted above. [Emphasis added.]

[ 121 ] The proposition that the double compensation to which the parties refer is not the double compensation of an individual participant also contradicts the trial judge’s conclusion concerning the purpose of s. 88(3): paras. 21, 54 and 98. My colleagues identify no palpable or overriding error in this regard.

B. Stipulations of the Contract

[ 122 ] It is true that the ASRA Program is primarily collective in nature, since its goal is to pay compensation for decreases in the income of insured businesses that is based on receipts and expenses calculated using a benchmark farm model. In other words, it does not necessarily cover each participant’s individual losses. It is also true that La Financière has some discretion as regards the data it can take into account for the general purposes of its calculations. Section 87 provides that La Financière can adjust and fix the [ translation ] “net annual income on the basis of statistical studies or of any other data it deems relevant”. However, that section does not authorize it to disregard the provisions of the ASRA Program, including the specific provision on how amounts received from other government agencies are to be taken into account.

[ 123 ] Section 88(3) gives La Financière the power to take into account [ translation ] “amounts to which a participant is entitled ” under a program of another granting organization. This means that La Financière cannot be penalized by a participant’s decision not to participate in such a program. On the other hand, La Financière may not penalize a participant for amounts to which he or she is not entitled owing to the inherent limits of those other programs. For example, it cannot take into account amounts to which a participant is not entitled owing to caps imposed by such programs. Whatever linkage method is chosen, La Financière must take the characteristics of the programs into account when amounts are granted under them to its participants on an individual basis. If it disregards the inherent limits of those programs and if the amounts attributed to participants bear no relation to the amounts they actually received, it is in breach of its contractual obligations.

[ 124 ] In the past, as to the Canadian Agricultural Income Stabilization program (“CAIS program”), the only other program that imposed caps, La Financière linked amounts individually. Section 88(4), which deals with the consideration of amounts obtained under the CAIS program, also refers to amounts to which a participant is entitled :

[ translation ]

88. The following items shall be taken into account in calculating annual receipts:

4° An amount that represents the amount to which the participant is entitled, for the insurance year or part of the insurance year, on the basis of deemed participation at a coverage level of 100% in the Canadian Agricultural Income Stabilization (CAIS) program divided by two thirds.

However, the amount to which the participant is entitled for the purposes of this calculation may not exceed two thirds of the difference between the participant’s production margin and the participant’s CAIS program reference margin.

[ 125 ] More importantly, by referring to a “participant”, s. 88(3) requires La Financière to take into account the amounts that were actually received. The term “participant” ( adhérent ) is defined in s. 2 of the ASRA Program as [ translation ] “a farming business, or any group of farming businesses that is recognized as eligible by the Program, that participates in the Program”. This definition refers to an individual participant in the ASRA Program, not to a “benchmark farm”, just as the definition of “annual receipts” ( recettes annuelles ), also in s. 2, refers to [ translation ] “compensation, subsidies or grants obtained during the year from government agencies”, not to notionally attributed amounts that bear no relation to the amounts paid directly to the participant.

[ 126 ] In this case, La Financière did not take the impact of the caps under the COPB and Kickstart programs into account, because the benchmark farm’s ANS were lower than those caps. If it had taken those caps into account, the difference between the amounts that were actually received and the attributed amounts would have been considerably smaller. Instead, it subtracted amounts in respect of some producers that represented up to 700.95 percent (Inter Agro inc.) of the amounts to which they were otherwise entitled under the ASRA Program. La Financière disregarded the amounts that had actually been received.

[ 127 ] The evidence also shows that, where amounts have been granted directly to a producer under another program in the past, La Financière has never used a linkage method — whether collective or individual — that had the effect of negating the compensation to which participants were entitled by attributing amounts that disregarded the inherent limits of those programs. La Financière’s choice in the appellants’ cases is therefore also inconsistent with its past practices.

[ 128 ] This was the conclusion reached by the trial judge, and I see no reason to intervene in this regard:

[ translation ] La Financière agricole argues that it has a discretion to decide what method to use in linking amounts. It is right. However, its discretion must be exercised in accordance with the terms of the program (or the conditions of the contract) and in a reasonable manner.

In this case, the linkage method chosen by La Financière agricole was not reasonable, because it was not consistent with the terms of the program that was in force in 2007 and because it was not fair.

By using the benchmark farm model to collectively link the assistance provided by the federal government under the COPB and AgriInvest Kickstart programs, La Financière agricole transposed the effect of the cap on ANS to the insurable volume and thereby indirectly reduced the insurance coverage. Producers whose ANS exceeded $450,000 under COPB and $3 million under AgriInvest Kickstart were penalized, as their compensation was significantly reduced.

La Financière agricole also attributed to participants that had received Kickstart benefits under the AgriInvest program the value of the contributions they would have had to pay to obtain the equivalent of those grants in ASRA compensation even though no contributions were required from them under the federal program.

The court is of the opinion that La Financière agricole did not exercise its discretion reasonably when it decided to link the amounts collectively. It should have linked them individually, because the collective approach created distortions that were too great and that unfairly penalized many participants.

Those distortions were foreseeable. The representatives of La Financière agricole who testified before the court admitted that the decision to link the amounts collectively was one of policy and that La Financière agricole had made the choice to favour certain producers to the detriment of others. In doing so, it breached its contractual and statutory obligations to the plaintiffs. It acted arbitrarily.

Some producers benefited from having the amounts linked collectively, as the reduction in their compensation was lower than the amounts they had received, but for others, the reduction was greater than the amounts they had received.

In the former case, participants benefited from double compensation, while in the latter, the participants were not fully compensated for the insured loss despite the terms of the contract.

When all is said and done, the linkage method chosen by La Financière agricole had the effect of creating two classes of participants, which is discriminatory. Moreover, in many cases, it did not contribute to avoiding double compensation, while in others, it prevented participants from receiving the compensation provided for in the contract, which is improper. [Emphasis added; paras. 161‑70.]

[ 129 ] What is problematic in this case is that La Financière’s action had the effect of overcompensating certain participants to the detriment of the others, that it disregarded the impact of the caps under the federal programs on the insurance coverage of participants and that the amounts it ultimately attributed bore no relation to the amounts that had actually been received, not that it chose a particular linkage method.

[ 130 ] Finally, the federal programs in question in this case can be distinguished from most such programs that have existed in the past, given that they involve amounts granted directly to producers and not a global amount. My intention is therefore to call into question not, as my colleagues suggest, most of the decisions made by La Financière over the past 15 years, but only the decision it made in this case.

[ 131 ] Before concluding, I should say a few words about the amendment La Financière made to s. 88(3) in November 2008, namely the addition of the following paragraph ((2009) 141 G.O. 1, 51, s. 21):

[ translation ] Those amounts shall be determined on the basis of the characteristics of the benchmark farm described in Table 4 unless La Financière agricole considers it appropriate to add the amounts actually received by each participant to the annual receipts, having regard to the specific terms of payment for each government initiative.

[ 132 ] According to the rules of interpretation, it must be presumed that that amendment served a useful purpose and gave La Financière a discretion it did not previously have: art. 1425 C.C.Q. If, however, as La Financière argues, the amendment was made simply to clarify the meaning of s. 88(3), this would confirm that there was an ambiguity. Any ambiguity that may have existed before the amendment was made must be resolved in favour of the adhering party, since this is a contract of adhesion: art. 1432 C.C.Q.

[ 133 ] Although it is true that the ASRA Program has 16,747 participants and that the interests of the participants may vary, they all expect to receive the fair compensation they are owed. As well, it would not be reasonable for all of them to expect to be favoured to the detriment of the others. In this case, to interpret the contract in favour of the “adhering party” simply means to interpret it such that La Financière cannot use notional amounts to deprive a participant of all or part of the compensation to which it is entitled.

III. Conclusion

[ 134 ] For these reasons, I would therefore allow the appeal in part.

[ 135 ] In my view, the trial judge’s award should be reduced by subtracting from it, as suggested by La Financière, an amount equal to the contributions the appellants would have had to pay in exchange for higher compensation.

[ 136 ] It would indeed be unfair for the compensation to be increased without taking into account the amount of the related contributions the participants would ordinarily have been required to pay had it not been for La Financière’s decision. The evidence adduced on this point is uncontradicted, and it indicates that, if the amounts had been linked individually, the appellants would, once the additional contributions were subtracted, have been entitled to $7,489,323.59, not $14,901,559.55.

Appeal dismissed, Côté J. dissenting in part.

Solicitors for the appellants: Beauvais Truchon, Québec; Paradis, Lemieux, Francis, Bedford, Quebec.

Solicitor for the respondent: La Financière agricole du Québec, Lévis.