Actuarial Notes

Canadian Annual Return

Study guide: Exam 6C

The Canadian Annual Return is the standardised regulatory filing every Canadian P&C insurer submits to OSFI and to provincial regulators after each year end. It contains the audited financial statements, extensive supporting exhibits (loss development, reinsurance, investments, business by province and class), the MCT capital calculation, and the Appointed Actuary's Statement of Actuarial Opinion.

  • What it is for. It is the regulator's primary window into an insurer, and it is standardised so that companies can be compared with one another and monitored over time. The CCIR and OSFI publish the instructions that govern its preparation.
  • Principal components:
  • The Quarterly Return is a reduced version filed each quarter, giving OSFI a more frequent read on capital and results.
  • The claims development exhibit is the most heavily used by analysts: it shows how each accident year's estimate has moved over successive valuations, so persistent adverse or favourable development becomes visible and cannot be concealed by a single year's release.
  • Public availability and the derived ratio sets — the MSA Ratios published by MSA Research from return data — are what make Canadian P&C financial analysis possible from outside a company.
  • The return was substantially rebuilt for IFRS 17: line items now follow Insurance Revenue, Insurance Service Expenses and Insurance Finance Income or Expenses rather than written and earned premium and incurred losses.

Reading the Development Exhibit An insurer's annual return shows, for accident year 2021, the estimated ultimate claims as reported at successive year ends:

\$142M (2021), \$148M (2022), \$157M (2023), \$169M (2024).

Surplus is \$210 million and total unpaid claims are \$480 million. What would a regulator or analyst conclude?

Answer Cumulative adverse development of \$27 million on one accident year — 19\% above the original estimate — and it is accelerating (+\$6M, +\$9M, +\$12M).

The inferences, in order of importance:

  1. The reserving process is biased low, not merely uncertain. Three consecutive movements in the same direction, growing in size, is not random error around an unbiased estimate.
  2. Other accident years are probably affected too. Whatever assumption is wrong — development factors, trend, a change in claims handling — applies across years, so the exhibit for 2022 and 2023 should be examined for the same pattern. If the whole book is 19\% light, the deficiency is roughly \$91 million against \$210 million of surplus.
  3. Pricing is contaminated. Rate indications built on understated loss estimates produce inadequate rates, so current business is being written at a loss — and under IFRS 17 that should be surfacing as onerous groups.
  4. The MCT is overstated to the extent liabilities are understated, so the reported capital ratio flatters the position.

What should have happened first. The Appointed Actuary performs a roll-forward or actual-versus-expected analysis at each valuation; three years of one-sided deviation should have triggered a change in method or assumptions by the second observation. An exhibit that shows this pattern while the opinion remained unqualified invites the regulator's obvious question.