Paul Carroll
Your latest economic outlook struck me as reasonably optimistic, despite the U.S. tariffs and our war against Iran, but the environment changes awfully fast these days. Where do you stand now?
Michel Léonard
I'm sorry to disappoint. We put out our Outlook a few weeks ago, and I've already turned more pessimistic. You know me — I lean optimistic by default. But here's what happened, and it all comes down to the Iran conflict:
When we put out our last quarterly estimates, everyone was talking ceasefire — and what it would mean for the P&C industry, replacement costs especially, and for the broader U.S. economy, inflation especially. Then the ceasefire evaporated, and the consensus swung back to the war grinding on. So we started repricing growth and inflation across the board — systemic, structural, economy-wide.
I say more pessimistic because I'm not ready to call a trend yet. You've heard me say it takes two quarters to make a trend — and we're not there. We won't have two clean quarters of data until late this year or early next, on what the Iran conflict has done to the economy, or is still doing by then. But we've clearly shifted from worrying mostly about GDP growth to worrying about price stability and jobs, too.
Paul Carroll
Oil prices haven’t risen as much as many of us expected, because various countries, especially China, have cut way back on consumption or dipped into strategic petroleum reserves. As a result, the oil shock hasn’t spread throughout the economy as much as it might have. But with the resumption of at least occasional fighting in Iran and the depletion of petroleum reserves and stockpiles for fertilizer and other oil-dependent products, are we reaching a breaking point where contagion can no longer be contained?
Michel Léonard
Yes, absolutely. We've been measuring how oil-price contagion spreads into the wider economy two ways: first, the damage already done, through existing PPI [produce price index] data; second, how it could unfold from here, through stress tests.
On what's already happening: higher oil prices have jumped the fence from energy into everything else — the contagion is real. Aluminum and petrochemicals, machinery and equipment, electronics and accessories, fertilizer manufacturing, plastic resins and materials — all up an average of 34% year-over-year over the last six months.
Paul Carroll
As the conflict drags on without a clear path to negotiations or a long-term solution, what do your stress tests say might happen to replacement costs in P&C insurance?
Michel Léonard
Our P/C replacement-cost stress tests ran three conflict durations — through end-2026, into 2027, into 2028 — and zeroed in on the three lines the Iran conflict is already hitting hardest on replacement costs: farm owners, commercial property, and personal and commercial auto. For those lines, we estimate replacement costs climb an average of 13% year-over-year if the war runs into 2027, and another 20% if it drags into 2028.
Paul Carroll
How should people use scenarios like this to plan?
Michel Léonard
In our last outlook, we said the whole point of stress tests is to build consensus around managing the severity of tail events — not their likelihood. Scenarios have to be credible, but not the most likely — and almost never are.
Let a stress-test conversation slide into political debate about how likely each scenario is, and it's dead on arrival. Pick credible scenarios, skip the likelihood argument, and go straight to impact severity — that's how you actually get people to consensus on a risk-mitigation strategy.
For P/C, that brings us to actuarial analysis — backward-looking by definition. Here, that's a feature, not a bug. You anchor to the data in the rearview mirror, then work with regulators on where reserving and rates should go. The stress tests are the headlights — you use them after, mostly to set reserving or policyholder surplus above what regulators require.
Now, if you ask which of the three is most likely, I'll say this may be the most impossible conflict to forecast I've seen in decades. Put a crystal ball in my hand instead of structured analysis, and I'd still bet on the one-year scenario — into 2027.
But I'll say it again: each scenario's exact probability is beside the point, as long as it's credible. What matters for our audience is that nobody can argue the war couldn't drag on another year or two. So my advice: reserve adequately for all three scenarios — in a way that's economically viable for your own book and how it actually behaves.
Paul Carroll
No matter what happens in terms of costs, the outlook for growth seems fairly positive. Is that correct?
Michel Léonard
The pressure point is what the Fed does. Real tell: the new chairman's very first communiqué ended with a standalone line — "The Committee will deliver price stability." That's just not how these things are written. I've never seen anything like it. It signals rate hikes for an inflationary environment, it locks the Fed in. And markets didn't like that line one bit when there was no rate increase at the next meeting.
I won't get into the dynamics between the new chairman and the president who appointed him — or what the president wants, which everyone knows. But if the Fed hikes based on today's inflation drivers, that's a basic policy mistake. Rate hikes work best on demand-driven inflation; this is supply-side, so raising rates buys you almost nothing.
As Greenspan used to say, rate hikes would be the wrong medicine for the disease. here This inflation is the White House's to cure — not the Fed's.
In my view, a 25 basis point hike in this geopolitical climate would land with several times the force it would in normal times. It would slam the brakes on growth — tipping GDP toward a contraction and pushing unemployment toward 5%.
Paul Carroll
A fair amount could change on tariff policy and other issues based on what happens in the November midterm elections. How do you see these political dynamics affecting the situation?
Michel Léonard
I'm so glad you brought that up, because the tariffs are back. It's August, and out of nowhere a few weeks ago we've got tariffs on Canada and others again. I didn't see that coming — I thought we'd closed that chapter. The only way they make sense right now is to change the subject, from Iran-conflict inflation to protectionism. But it's not all performative — tariffs are increasing government revenues, real money coming in, more than expected back in 2022.
Tariffs are inflationary by design. They won't do affordability any favors — and insurance affordability least of all.
The midterms and the rest of the DC noise you're pointing to — that's what tips the one- or two-year scenario.
The one-year scenario gets more likely if Democrats lead the polls into the midterms and take at least one chamber of Congress — this could cause a replay of 1980, when Reagan was already offering a better deal to the Iranians before he'd even won the presidency.
The two-year scenario gets more likely if the Democrats don't win back a chamber. You'd have a lame-duck president who'd probably want to end the war he started on his own watch — or, again, we could have a rerun of 1980.
Paul Carroll
What happens if the Iranians show for the long term that they can keep the Strait of Hormuz mostly closed and that the U.S. can’t protect our longtime allies in the Middle East as much as we and they thought we could?
Michel Léonard
That changes the whole ballgame — the repercussions would run in every direction. But now I'd be trading structured analysis for the crystal ball again, so I'll stop there.
Paul Carroll
We’ve covered a lot. How would you summarize?
Michel Léonard
First, prices are already spiking — and not evenly; some costs aren't behaving the way you'd expect. Check the actual BLS and BEA data. That part matters. Second, P/C replacement costs are climbing, too, hardest in farm owners, homeowners, and commercial property.
Finally — and I'll end on optimism, like I promised — I think this slowdown is short-lived. As long as the Fed keeps its hands off rates.
And if you’re a Triple-I member, don’t forget that you can reach out to us directly. Membership has its privileges.
Paul Carroll
Thanks, Michel. As insightful as always.
