IFRS17 – Twenty Third Floor

IFRS17 consistency – VFA vs GMM

David Kirk — Thu, 16 May 2024 13:11:19 +0000

Your experience with IFRS17 over the past couple of years likely depends on the insurers you’ve been closest to—whether they’re predominantly using PAA, GMM, or VFA approaches.

The differences between the General Model (GMM) and the Variable Fee Approach (VFA) can be substantial. VFA does away with the awkwardness of locked-in rates on CSM and the resulting accounting mismatches as interest rates drift up and down over time. This can significantly affect the presentation of profitability and financial position.

While IFRS17 has removed some inconsistencies, it has introduced others, including multiple definitions of operating profit. I’ve seen references to at least 10 variations, which is unhelpful.

During a recent InsuranceERM conference in London, only 30% of the audience felt that IFRS17 had led to improved comparability and consistency.

I don’t believe comparability within a single country has necessarily improved, but IFRS17 may have better aligned reporting practices across different countries. However, opinions on how much consistency has improved may vary depending on your perspective.

If you zoom our far enough, all the differences disappear. But up close, the departures are key.

Frictional cost and tax

David Kirk — Wed, 15 May 2024 15:33:49 +0000

There are many reasons to doubt the perfect applicability of the 6% cost of capital rate used in South Africa for the solvency Risk Margin calculation.

Not least of which is the decrease to the rate in Europe and in the UK.

However, if we borrow ideas from Embedded Value (TEV/EEV or MCEV) and look at the components of…

A) a required premium or return for risk (2% to 6% or even higher depending who you ask); and
B) a frictional cost for taxes and shareholder investment expenses

…it becomes hard to justify a rate much lower than 6% in South Africa.

One reason for the difference from the conclusion in Europe? The absolute level of our interest rates and the additional tax drag on that. (Incidentally, this is the same reason it’s hard to make a real return outside of retirement savings vehicles and Tax Free accounts, and also why it’s more tax efficient to invest in hard currencies.)

Keep an eye on ‘Frictional Costs’—a term that’s likely to become more relevant as EV reporting evolves and MCEV ideas come alive again. This could easily be 2.5% to 3.5%.

Here’s an illustration to ponder. Your results may vary based on assumptions.

IFRS17 may not kill off EV

David Kirk — Sat, 11 May 2024 15:39:48 +0000

Will IFRS17 kill off Embedded Value (EV) reporting in Africa?

Or will it finally bring Market Consistent Embedded Value (MCEV) to life?

I gave a presentation at the Life Assurance Seminar 15 years ago on MCEV. It took off in the UK but didn’t become popular in South Africa. That might be changing.

Some insurers have already stopped EV reporting altogether. This has some pretty unattractive implications for lines of business where using solvency-based measures with short contract boundaries distorts value.

One of the simpler (and most useful) ways to report EV figures in an IFRS17 world is to adopt MCEV principles and pull most of the relevant figures out of existing IFRS17 reporting. If you are comfortable that your Risk Adjustment is appropriate, adjusting CSM for tax, non-attributable expenses, and frictional costs can get you to an acceptable MCEV.

Other changes are still required for contract boundary extensions and non-insurance business. Will insurers have appetite to value these on a directly market consistent basis, or will these non market consistent values be aggregated along with purer MCEV for life insurance lines? (There’s no fundamental problem here – value is value regardless of the method.)

Insurers have not settled on a single reporting framework. Internal measures are not even always consistent with external reporting. We absolutely need consistent, comparable, rational measures. Not least because with Value of New Business (VNB) margins under pressure almost everywhere, and analysts increasingly asking pointed questions around onerous contract (under IFRS17), an accurate and reliable measure of new business value that everyone agrees to is critical.

A wild idea

David Kirk — Wed, 20 Mar 2024 07:35:00 +0000

I’ve been brewing a wild idea for a while.

Insurance regulations weren’t written with IFRS17 in mind. This causes some head scratching when it comes to premium volume measure for non-life insurance, but common sense gets you to the right answer without much trouble. Those who say otherwise seem to be looking for problems where none exist.

I have been pondering whether IFRS17 makes life interesting for microinsurers given the wording of FSM2 “Valuation of Assets, Liabilities and Eligible Own Funds” for microinsurers (issued by the PA). The interpretation and application challenges actually predate IFRS17. FSM2 makes some silent and unlikely assumptions around treatment of premium debtors for typical microinsurance business. More on that in a future article.

IFRS17 does make life interesting (in the worst meaning of the word) for microinsurers, in that they must all apply IFRS17 to their insurance contracts. There’s no reason not to apply the Premium Allocation Approach given restrictions on policy term – and this simplifies many of the calculations significantly. Whether the audit firms looking at microinsurers understand IFRS17 or the required disclosures is an important quite separate topic, but one which must be resolved independent of the prudential reporting basis itself.

So here’s a wild idea. Why not drop FSM2 altogether and align the prudential balance sheet with the IFRS one?