With temperature records melting under the European heatwave, we call on the reinsurance industry to cool off on manual submission data preparation.
And where better to learn, than from a brief history of weather data collection? Read on for mountain memoirs, ocean odes and sky stories...
...and don't forget to follow Supercede | The Reinsurance Platform, for more like this!
More than one hundred years ago, the summit of the UK’s tallest mountain was inhabited all-year-round by a team of scientists, together sustaining a manual and repetitive effort to capture accurate mountain weather data. The aspirations were noble, but the methods were far from scalable.
The process attracted substantial expenses and made for a soul-destroying experience, as the handpicked team of scientists weathered the daily drudgery of data collection at 4,400 feet. As the archivists put it:
“Life at the summit was often difficult, especially during the long, cold and dark days of winter, when staff battled freezing winds, torrential rains and solitude.”
One can only imagine how much more difficult it would have been to scale this effort in landscapes where mountains are nearer to 30,000 feet in height – or worse still, to replicate it in skyscrapers filled with reinsurance professionals, where the ‘cold dark days of winter’ take the form of 1/1 renewals.
By 1904, after twenty-one years of this experimental tragicomedy, it was recognised by all that a less manual data collection solution was required for meteorology and climatology to prosper. In its unassailable laboriousness, the cost of the mountain mission was deemed unsustainable, and the scientists were gratefully brought back down to earth.
All very well for the climate scientists, who made it back down the mountain more than a century ago. But for the reinsurance professionals still up there, stuck in a perpetual repetition of submission data groundhog day, how long will it take to find a safe path down?
Much like the reinsurance industry, weather forecasting can trace its roots to a shared public and private interest in the safety of life and property at sea.
The first ‘Met Office’ – the UK’s national weather service, or meteorological office – was established in 1854 by none other than Robert Fitzroy, Captain of the HMS Beagle and accordingly, the travelling companion of Charles Darwin. Already at the time, you could follow the course of the HMS Beagle in Lloyd’s List.
Within its first decade, the new Met Office published both its famous Shipping Forecast and the first ever public weather forecast. Previously, only a few private forecasts had been available, with limited coverage. However, thanks to the data collected, aggregated and validated by Fitzroy’s team, through a scientifically consistent method – and the ability to share data rapidly via the electric telegraph – the Times newspaper began in 1861 to publish the Met Office’s forecasts regularly.
The forecasts covered the length and breadth of the UK alongside the islands of Jersey, Guernsey and parts of France, in uniform fashion: something most insurance companies still struggle to achieve, due to differing datasets across their various fiefdoms, acquisitions and trading entities. And, with the value of the outputs being put to increasingly good use by the populace, investment in getting regular, reliable data soared.
Before anyone knew it, weather visionaries had found a way to finance a living hellscape for scientists atop Ben Nevis, as described in the noble misadventure set out at the fore of this article. There, as with reinsurance pricing actuaries pulling late ones through renewal periods, experts were deemed necessary to ensure the gruelling, janitorial work was done properly. But due to the months-long process of manual data collection and verification, barely an ounce of their intellect could be deployed: less still in a timely enough fashion to be of material use to anyone.
Fortunately, at least for the weather industry, things were not always to remain this way. Standard means of data collection were enabled with the advent of computers, supercomputers and super-duper-computers; enabling weather information to be rapidly calculated and distributed live, to a vast array of information consumers around the world. In contrast to our own industry, the solid fundamentals of weather datasets – underpinned by reliable, consistent methods of generating the critical cohort of weather exhibits – have enabled innovative forecasters to overlay inputs from new data sources, such as satellites and internet-of-things devices, with enviable dexterity.
In the land of reinsurance submission data, we swap standard weather metrics, such as temperature, wind speed, humidity and air quality for our own equivalents: premium estimates, rate changes, loss runs and bordereaux. In place of climate charts by location, we seek limit and attachment profiles by occupancy. And for both worlds, effective year-on-year analysis has proven vital to enabling meaningful discussions, whether the topic under debate is climate change or a rate decrease.
But reinsurance data is struggling to keep pace. Not only is the weather forecast now exceptionally accurate in its assertions – pinpointing the forecast for the UK’s record 40.3C temperature this week – it also affords us the possibility to drill down to even the most precise locations imaginable. As it happens, I’m currently in Llanaelhaearn, which I am informed with a quick weather search is a degree cooler than nearby Pwllheli (maybe I'll try to pronounce these on The Reinsurance Podcast sometime).
Meanwhile, the teams dealing with reinsurance data would, as things stand, give an arm and a leg for either accuracy or precision: not in their wildest fantasies had they considered the prospect of achieving both. Some details, they can get at the section level but not at the subsection level, and vice versa; other numbers, they simply don't trust.
Fortunately, we see a change in weather on the horizon. And no, we don’t have our heads in the clouds: we're here to help you assemble your reinsurance submission data in the cloud.
By the same name as our mission, we show cedents, brokers and reinsurers the way to supercede the sorry state of affairs that has beset reinsurance submission data for many years too many. We believe in a better way to deal with reinsurance data: because in the age of satellites and supercomputers, standing on a mountain in a snowstorm just won’t cut it anymore.
If you're up there, clinging to the summit, and miles away from what’s happening on the ground, trust me: it’s only going to get worse. The winters will only get harsher, the data will only come under more scrutiny. After all, these are the days of space rockets, not crampons.
It’s time to climb down from the manual reinsurance data mountain.
It’s time to .
Don't forget to follow Supercede | The Reinsurance Platform, for more like this!
