As defined by IAS 37, a provision is a liability of uncertain timing or amount. To understand this definition, one must know the definition of liability.
A liability is:
- a present obligation arising from past events, and
- the settlement of which is expected to result in outflow of resources embodying economic benefits.
A provision will be recorded if the above definitions are met and a reliable estimate of the amount of the obligation can be made. The term obligation includes contractual as well as constructive obligations.
Contractual obligation is an obligation that arises out of a contract or agreement whether in written or verbal form. Contractual obligations are legally enforceable. For instance, an obligation to pay for the goods purchased is a contractual obligation, as at the time of purchase, the buyer agreed the price with the seller.
Constructive obligation is an obligation that arises out of an entity’s past practices, published policies or statements, casting a valid expectation in the minds of third parties that the entity will perform a certain task. Constructive obligations may not be legally enforceable. For instance, a store has a regular practice for several years that it accepts purchase returns within one week and refunds the amount to its customers. The refund policy is not explicitly agreed with the customers, so the store is not legally bound to refund the amount. However, the store’s practice of accepting purchase returns for several years has developed a valid expectation in the minds of the customers that if required, these goods can be returned, and the store will refund the amount. Therefore, this is a type of constructive obligation of the store.
Estimating the amount of provision
A provision is recorded if it meets the definition of provision and the amount of obligation can be estimated reliably. As the timing and amounts are uncertain, so it is obvious that some estimation is required. The following two approaches can be used.
Most likely outcome
If the event under consideration has only two or three possible outcomes, the best estimate would be to consider only the most likely outcome to determine the amount of provision. For instance, provision for a potential liability of an entity arising from a court case that can be settled either in the favor of the entity or against it should be measured based on the most likely outcome.
Expected value or probability weighted value
If the event under consideration has numerous possible outcomes, the best estimate would be to consider the expected value of the obligation which is a probability weighted value of all the possible outcomes. For instance, provision for warranty is measured by calculating the expected value of cost of repairs based on an entity’s past experience.
Let’s go through an example of provision.
Example – Provision for refund
An electronics store has a policy of accepting purchase returns and refunding the full amount if the customer returns the goods within one month of the purchase. Past history of returns is evaluated by the store and following average trends have been noticed.
- Product A: 5% of the sold goods are returned by the customers
- Product B: 15% of the sold goods are returned by the customers
- Product C: 2% of the sold goods are returned by the customers
The store is preparing its monthly financial statements, where the total sales of $100,000 are made during the month. Product-wise breakdown of sales is given below.
- Product A: $60,000
- Product B: $30,000
- Product C: $10,000
Calculate the provision for refunds required in the following two scenarios:
- The electronic store explicitly agrees the refund policy on the sale invoices.
- There is no explicit agreement, but the customary business practice of accepting purchase returns within one month is well known to the customers.
Let’s start with the discussion whether there is any provision required or not. A provision is a liability of uncertain timing and amount, whereas it includes both contractual and constructive obligations.
In scenario A, there is a contractual obligation to accept the purchase returns and based on the historical data, a reliable estimate of the provision can be made, therefore, provision for refund is to be recorded.
In scenario B, there is a constructive obligation to accept the purchase returns as the well-known practice of accepting purchase returns has developed a valid expectations among the customers that they will be refunded by the store. Furthermore, based on the historical data, a reliable estimate of the provision can be made, therefore, provision for refund is to be recorded.
Calculation of provision for refund will be same in both scenarios.
Best estimate of the provision for refund is calculated below.
Product A: Sales x Probability of refund = $60,000 x 5% = $3,000
Product B: Sales x Probability of refund = $30,000 x 15% = $4,500
Product C: Sales x Probability of refund = $10,000 x 2% = $2,000
Total provision for refund = ($3,000 + $4,500 + $2,000) = $9,500