Combined ratio rose 4.9pp to 90.3% as large events drove underwriting lower

Large-event claims expense was $9.3m versus $2.8m, taking the large-event claims ratio to 5.6% from 1.8%, while the claims ratio excluding large events also rose to 48.2% from 44.6%. The expense ratio improved to 36.5% from 39.0%, partly offsetting.

Underwriting profit fell to $16.2m from $22.9m, net investment income dropped to $0.7m from $2.2m, and NPAT was $11.5m versus $14.4m. Revenue edged up to $203.5m from $199.8m; specific growth rates for revenue, PBT and NPAT are flagged as not analytically comparable on a like-for-like basis given basis changes in the historical series.

Operating cash flow jumped to $58.8m from $4.0m, lifting pre-lease free cash flow to $58.4m — above Annolyse's historical baseline (4-period mean $28.9m, range $16.4m–$47.8m). Tower restarted dividends with a 2.5c interim, its first in five years, with the New Zealand parent's solvency ratio at 309% ($97m above the minimum).

The claims ratio excluding large events rose 3.6pp to 48.2%, alongside the 3.8pp lift in the large-event ratio. Underwriting profit dropped $6.7m even as net earned premium grew to $167.1m from $159.4m. This matters because top-line momentum is not being matched by claims discipline, and the BAU element cannot be explained away by weather.

Underlying NPAT shows attritional pressure. Management's underlying NPAT before large events was $11.3m versus $16.9m — a $5.6m fall that strips out the catastrophe distortion. The implication is that expense-ratio gains from the digital-first program are being eroded by claims, not amplified by them.

The dividend restart is funded by capital, not current earnings. Solvency of 309% with $97m above the minimum provides clear headroom for the 2.5c interim despite the underwriting setback. Resumption of distributions therefore reads as a capital-management signal rather than an earnings-quality endorsement.

With HY21 reported NPAT of $11.5m and underlying NPAT of $11.3m, hitting the midpoint requires roughly $14.5m in the second half — materially above the first-half run rate. The FY20 second half delivered negative implied NPAT, so prior seasonality is not a reliable shape guide.

The implied 2H lift assumes either fewer large events than the $9.3m already consumed, BAU claims normalising back below 48%, or further expense leverage. None of these are confirmed by this release, and the FY21 large-event budget has already largely been spent.

The 2.5pp fall in the expense ratio looks structural and consistent with the stated cost program, but the 3.6pp rise in the BAU claims ratio overwhelmed it. The fact that underlying NPAT (which excludes large events) fell to $11.3m from $16.9m indicates the deterioration is not solely catastrophe-driven and may carry into FY22.

Cash flow quality requires care given the insurance context. The OCF jump to $58.8m, and pre-lease FCF of $58.4m, sit well above Annolyse's historical baseline — for an insurer this typically reflects premium-receipt and claims-payment timing rather than recurring earnings power. Cash and equivalents actually fell to $85.1m from $101.4m despite the OCF surge, so the working-capital uplift was deployed rather than accumulated. Capex of $0.5m versus $1.8m flatters short-term cash but raises a question about the durability of that reinvestment cadence.

This briefing cannot assess the adequacy of claims reserves or the catastrophe exposure profile underlying the FY21 large-event allowance.