Subject: Variable Insurance Product XBRL Taxonomy - S7-23-18
From: Anonymous

Mar. 31, 2020

Dear SEC: 

I do not agree with the new XBRL requirements for variable products. These products are too bespoke for useful comparison purposes. 

Variable life policies terms are based on very specific underwriting requirements and the two most important pieces of the policies would not be picked up; your face amount and your fee. All other disclosures are too minor to be meaningful. 

Variable annuity contracts are far more complex than mutual funds and ETFs. In a mutual fund, all holders of a class have the same deal. In a variable annuity, every investor can select their own features, the features may vary by age, gender, state, time of purchase, and marital status, and the values of those features change frequently. You cannot simply compare your current variable annuity with a new one, or even compare 2 competing variable annuities. Looking at the most popular variable annuity on the market today, I believe that one contract would have more than 2,800 tags. This is too much information to be useful. I am not sure the SEC appreciated this complexity when they proposed to tag just a few items in a prospectus. Has the Commission (those 5 people) actually seen what the data looks like and every tried using it? Has the staff sat there with a stopwatch and tried to tag a typical variable annuity prospectus? Has any vendor come to you and said I plan to sue your data to create an insurance product comparison tool? If they did, did they suggest you tag these items in this fashion? Your estimates of the usefulness and costs are way off. XBRL is pushed because it is a good talking point, and not because it will actually provide value to investors. 

You will also find that there is diversity in practice as to how insurance companies structure their disclosures and arrange their benefits. That also may make any comparison difficult. Some insurers describe each rider in a serial fashion (2 to 7 pages per benefit), and others describe the common elements together. Either way you will see very large block text that is not particularly useful for a side-by-side comparison. 

In the fund space, XBRL has not been a worthy endeavor. You can provide no evidence of any non-government agency effectively using this data. Instead, the market uses Morningstar, Lipper, Bloomberg, and CRSP. These organizations have been providing fund information long before the XBRL requirement and if the SEC stopped that requirement, they would continue to produce that same data. It is also worth noting that XBRL did not drive down the costs of these services. All this effort is doing is to enrich filing agents and improve the margins of data aggregators. 

I support XBRL in the operating company space because that information is necessary to quickly understand the value of your investment. In the fund and variable insurance product space where there is the concept of NAV, live tagged data is just not necessary. 

As an alternative, many states require certain sales agents to prepare a document that compared the old and new annuity involved in an exchange offer. If anything, you should copy those simplific requirements such that tagged data could provide a meaningful short-form comparison of two annuities. 

I apologize for writing this anonymously, but my employer, a large insurance company, would not otherwise permit me to send this message. 



P.S. This taxonomy was very difficult to find on I am not sure this was a good faith effort at eliciting public comment. The proposing and adopting release should have explicitly stated what should be tagged, how it would be tagged, and the level of detail.