I grew up thinking that actuarial science was one of the most advanced things you could do with math.
This was before AI was really a thing and before anyone told me what software engineering salaries were like out in the wild.
But...I didn't know any of this in high school so I took some actuary tests since they were a fun way of learning probability.
I have spent the last few days learning about what actuaries do and how insurance is priced. It's a remarkably deep rabbit hole, that seems almost untouched by 21st century technology.
I'll summarize a bit of this exploration, breaking up the big observations into strange dichotomies.
Thesis: Actuaries are tasked with pricing risk.
Antithesis: Actuaries are EXTREMELY risk-averse people.
There is very little well-produced media on the internet about what actuaries do every day and how they think about the world. This is probably because (a) there are only 25k actuaries in the US (there are 10x the # of software engineers), and (b) I would bet the profession is full of people uninterested in sharing their lives on the internet.
I have skimmed such blogs/videos/etc. It was entertaining, and I recommend others dive into this rabbit hole. The video above is good starting place.
It is clear that the personality types of actuaries are those that like clear-cut career advancement. Salary bands for actuaries are very predictable based on how many exams one can pass. Most actuaries believe passing a test every six months is ambitious but possible. Most firms require passing at least two exams before full-time employment.
After 6 exams are passed, one earns the designation of Associate of the Society of Actuaries (ASA). After a few more tests one earns the designation of Fellow (FSA).
Synthesis:
Maybe it makes sense to have people inclined to obsess over minimizing risk also be the people that price risk? Insurance is a conservative financial product after all.
Maybe it is important to be able to view the world in an objective and hierarchical way in order to consider risks logically.
BUT part of me still thinks there's no way risk-averse people are fundamentally able to understand and price risk. Not everything is an emotionless random variable, and much of what insurance should do is try to design creative ways to make people and property safer. Only risk-takers and creatives would see insurance this way, never actuaries.
Thesis: Actuaries use advanced math
At least hypothetically, the math actuaries are exposed to and must know to be licensed includes very advanced financial modeling. The math below, though not crazy hard, is something beyond what I have to contend with on a daily basis. This is also only from the first Exam P.
If you consider HFT or PE modeling to be sophisticated, you must at least acknowledge that actuarial science should be at least as sophisticated in principle.
HFT and PE modeling typically just model the price of a small range of assets (equities) whereas actuarial modeling goes much deeper and puts a price on health, life, and all kinds of property.
Antithesis: Actuaries consider VBA to be a sophisticated tool
Most actuarial models are written in Excel, still. Even billion dollar companies with dozens of full-time actuaries crack out VBA when the going gets tough.
The mindset of the industry is that better models come from better Excel skills. To get better Excel skills, you need more training and experience, obviously.
Work is almost never modularized. Though there exists a small open-source community of R and Python packages, most actuarial work is done with internal data-cleaning pipelines and with Excel files. It is not clear how collaboration works, but most actuaries appear not to know how to use Github.
Synthesis: actuaries do not see themselves as software engineers, and think the concept of open-source tools exposes too much intellectual property of the firm. Much of their day-to-day work is data cleaning, data visualization, and statistical analysis -- fields that have plenty of open-source tools available but are mostly unused by actuaries who prefer to stick in the Microsoft tools ecosystem (Excel / Powerpoint / Word).
Yes, I mean Microsoft Word. Many rate-filing documents are created as Word documents and submitted in PDF form (as official submissions).
Thesis: Getting credentialed in the insurance business is a long process.
Getting credentialed to become an insurance broker or claims administrator is very complicated and requires a lot of careful steps.
Most people get a lawyer to help manage all of the various filings, that vary state-by-state, required to become/stay allowed to sell insurance and manage insurance claims.
I had to pass two tests (neither hard but both of which required memorizing many different facts) and get fingerprinted.
There are law firms that specialize in managing just the insurance credentialing process, and get good business doing so.
Not having proper licensing is a huge risk to companies, and has taken down titans like Zenefits.
Synthesis: credentialing has killed the insurance industry. It is tedious but not difficult to become a fully-functioned player in insurance, whether as an actuary, claims adjuster, or an insurance broker. Hence, the industry attracts people that like easy, tedious things.
This is the same situation that has happened in medicine.
Antithesis: Insurance credentialing feels like filling out a first grade homework assignment.
One example: the official training for being able to use SERFF, a required platform to submit rate filings for new insurance products in a state, is...
Watch a video
Download the quiz (below) and complete
Submit the completed quiz (below) to your Product Specialist or the SERFF Marketing and Implementation team at serffmktg@naic.org (from https://www.serff.com/serff_new_user.htm)
The quiz is literally just a 10 question multiple-choice test with required score 6/10!
Conclusion: It's time to refactor this industry.
Let's delete some code. ;)