Reading

A French economic bilan depends on what it counts

Read a French economic report that changes its accounting boundary, separating aggregate gain, unequal distribution and omitted external costs.

In short

A bilan is meaningful only with its accounting boundary. Here the internal total is +20; including eight units of access spending gives +12. Neither figure proves that every site gains or that every external cost has been counted.

Orange counters sit inside a gold open ring; gold and teal counters occupy wider areas around it.
Changing the accounting boundary changes which costs enter the total. Illustration: Atelier

C2 · Accounting scope

Le périmètre du bilan

Le bilan du réseau annonce un gain de vingt unités sur l’année. Ce total additionne les résultats des trois sites, mais ne retranche1 pas les dépenses engagées par les communes pour adapter leurs accès. Il décrit donc un résultat interne au réseau, et non un bénéfice net pour tous les acteurs concernés.

La répartition du gain est également inégale. Le premier site gagne quinze unités, le deuxième sept, tandis que le troisième en perd deux. Le total positif n’empêche ainsi ni une perte locale ni une concentration de l’amélioration. Un commentaire qui annoncerait une progression de chaque site serait incompatible avec les chiffres fournis.

Un second bilan élargit le périmètre aux dépenses d’accès. Il estime celles-ci à huit unités pour la même période. Dans ce cadre précis, le solde2 devient douze. Cette soustraction n’intègre toutefois ni les effets environnementaux ni le temps supplémentaire consacré par certains usagers à leurs déplacements. Aucune valeur chiffrée n’est fournie pour ces deux dimensions.

Les auteurs ne proposent donc pas de remplacer un total vrai par un autre qui serait seul légitime. Ils demandent que chaque bilan annonce son périmètre, ses bénéficiaires3 et ses omissions. Le choix d’un indicateur dépend de la question posée. Pour décider d’un nouvel investissement, il faudrait encore expliquer le poids accordé à la distribution des gains et aux coûts qui restent hors du calcul.

Read it in English

The network’s report announces a gain of twenty units over the year. This total adds up the results of the three sites but does not subtract the spending incurred by the municipalities to adapt their access facilities. It therefore describes a result internal to the network, rather than a net benefit for every actor concerned.

The distribution of the gain is also unequal. The first site gains fifteen units, the second seven, while the third loses two. A positive total therefore does not rule out a local loss or a concentration of the improvement. A commentary announcing progress at every site would be incompatible with the figures provided.

A second report broadens the accounting boundary to include access spending. It estimates this at eight units for the same period. Within this specific framework, the balance becomes twelve. This subtraction still includes neither environmental effects nor the extra time some users spend travelling. No numerical value is provided for these two dimensions.

The authors are therefore not proposing to replace one true total with another that alone would be legitimate. They ask each report to state its accounting boundary, beneficiaries and omissions. The choice of an indicator depends on the question asked. Before deciding on new investment, it would still be necessary to explain the weight given to the distribution of gains and to costs left outside the calculation.

Question 1No account needed
Does +20 establish all sites gained?

Choose one answer.

Answer and explanation

No. The third loses two units.

Question 2No account needed
What changes +20 to +12?

Choose one answer.

Answer and explanation

Eight units of access spending. The second boundary includes that expense.

Question 3No account needed
Are all environmental costs priced?

Choose one answer.

Answer and explanation

No. No numerical value is supplied.

A faithful summary can preserve several correct totals under different definitions. The disagreement may concern which question each total answers and which criterion should govern a decision, rather than an arithmetic error.