To properly calibrate the Solvency II model for property-casualty companies, CEIOPS, the European regulator has requested premium, losses and loss triangles by line of business.
The data call is here, including an explanatory note, and the data is due with the rest of QIS5, in November. The request seems pretty simple by U.S. standards – gross loss and premium data, gross data with catastrophes excluded, net data with catastrophes excluded. All data is by accident year.
In the U.S., much of this is compiled annually, in the Annual Statement. In Europe, it’s a data call – and one that’s fairly late in the process, by my watch.
And that’s what has always confused me about Solvency II, sitting here in the States. The data requested is fairly simple. And it seems as if it’s difficult to obtain. Yet the entire continent is about to set capital standards based on this threadbare data.
And there’s little industry-wide data, as far as I know. There’s no statistical firms like ISO gathering data from many companies. There’s no simple aggregation of industry results, as you’d see in Best’s Aggregates and Averages. No excess loss triangles to compare with what the Reinsurance Association of America pulls together.
In fact, actuaries in the U.S. joke about the European lack of quality data. One consultant told me he was amazed at the intricate mathematical sophistication of European actuarial science, all applied to crap data. The result could hardly be considered technical guidance.
And I’ve heard the joke go the other way: European actuaries laughing how Americans have too much data to do any real math.
Yet Europe is leading the way in terms of actuarial sophistication. Solvency II is more technical than anything the NAIC is likely to approve. European actuarial research is certainly the equal of that in the United States, and some would argue it is superior.
So what gives? How does European property-casualty insurance operate, given the lack of industry data?