Tuesday, May 1, 2012

Adapt and Die: The Hidden Costs of Innovation


Quick: what’s one of the top five most common causes of insurer insolvency? If you’ve taken Exam 6, hopefully you remember that “rapid growth” is on that list, which seems strange at first. Growth is what investors come to market for. We treat it as a fundamental measure of economic health, whether it’s growth in the S&P 500 index or growth in GDP. For the insurer, however, idiosyncrasies of the insurance product turn growth into a mixed-blessing. Premium growth can be a sign that management is attempting to pay existing liabilities with new premium dollars, an unstable situation that could lead to insolvency. Growth is therefore not always a blessing in the insurance market.

Even for a non-insurance product, though, growth can turn out to be a curse. While companies must innovate and improve their products in order to stay competitive, fundamentally changing a product can ultimately cause a loss of market share if existing customers don’t take to the new product. Whenever a company tackles a radical new path and succeeds, analysts praise the bold vision. More frequently, however, you hear analysts declaring that the change was too bold, that it alienated the existing customer base, and that, in the long run, growth was not achieved. Product shifts have high rewards, but also high degrees of risk.

Two ongoing product shifts come to mind: Facebook and the Casualty Actuarial Society. Facebook pushes major changes on its users so frequently that the company’s name has become synonymous with unnecessary and unwanted “improvements.” As a less extreme example, the CAS constantly keeps two items on its agenda: solidifying the current base of actuarial services, and expanding the actuary’s role into new areas.

Facebook is a particularly good example of the risk posed by a changing product, because a social networking site, of whatever kind, can really only be one product at a time (at least, so it appears thus far). You’re either a dating site, or a “find friends from high school” site, or a business networking site, or a place to post pictures of cats. Over its history, Facebook has tried to be nearly all of these things, but at any given point in time, it has only been one.

At the start, Facebook was exactly what it sounds like: a facebook. In fact, at the time, it was referred to as “the Facebook,” and that was it. It wasn’t a verb or a place where things happened. It was a virtual Rolodex and little else. Being “Facebook friends” with someone had no moral significance. I ended up “friends” with every person I went to high school with, essentially just to keep track of where everyone was going to college. It had all the personal significance of a PDA.

Then came “poking” and “messaging.” I still don’t know what poking is about, but messaging quickly became cruising for dates if you friended any “friends of friends” (at least if you were foolish enough to list yourself as “single”). Feeling like I’d been tricked into joining an online dating site, I started avoiding the site, checking in only when I needed to find someone’s contact information, which was fine by me, given that the whole reason I joined was to absolve myself of the responsibility of keeping an address book.

The whole paradigm shifted, however, when “status updates” and “News Feed” were introduced. Facebook stopped being about enabling individuals to connect when they felt like it and instead became a platform for individuals to inject themselves into the spheres of others, almost forcibly. The Rolodex became a forum combined with Reddit, and that’s fine, too. But in executing this change, the original purpose was lost. I didn’t want to send such personal information, whether it’s where I’m having dinner or who I’m (not) voting for, to every person in my Rolodex. Further News Feed subjected me to similar information about others, information I was most often not eager to see (there are only so many pictures of food I can take).

This is when “Facebook purges” began. I couldn’t be Facebook friends with just anyone if I was going to be sharing photos and personal thoughts with all of them. It started with “no friends from work,” expanded to “no friends who creep me out,” and ultimately rose to the level of “no friends who cannot spell.” The set of Facebook friends contracted and contracted, shrinking closer and closer, until finally it became identical to the set of actual friends.

The original Rolodex model is completely gone, but in its place is a new product. At the end of it all, did Facebook grow? It would be sacrilege to answer “no.” The degree of investor confidence in Facebook is astounding. However if we take the limited view of Facebook’s original product, we have to answer “no.” The Facebook I joined before heading off to college is in bankruptcy, and has been for some time. This isn’t a problem if the new Facebook creates more value than the old Facebook (and it probably does), but it should give us pause to see such a radical shift. The kind of 180 that Facebook has made is not common. From a production perspective, it’s comparable to Coca-Cola ending production of regular Coke after introducing Diet Coke. From a risk perspective, it’s comparable to starting a totally new company, with a little bit of risk mitigation due to the inertia that will keep existing customers around temporarily.* It’s certainly not the obviously safe bet most investors seem to think it is.

For now, Facebook sits in this new equilibrium as a place where actual friends interact in ways that friends normally would, only faster and more easily. But Timeline is coming, and that means the product is changing once again. If the past is any indicator, that means the current model will soon have to go, in which case anyone joining the upcoming IPO will be investing not in a tried-and-true product with known demand, but in yet another speculative product, the market for which is unknown. Add to this growing privacy concerns and employers demanding access to employees’ Facebook accounts, and I see a very uncertain future.

As for the CAS, as I mentioned earlier, the organization is constantly looking for ways to grow the actuarial market. Most recently the CAS started offering the Chartered Enterprise Risk Analyst credential. While still an actuarial credential, CERA does expand the actuary’s focus into broader business concerns. The CAS’s interest in these general business risks is evident in the exam process, where Enterprise Risk Management topics are covered through papers written by non-actuaries and published in major business journals.

ERM certainly seems like a blossoming field, but there is still a trade-off to pursuing it. What more traditionally actuarial papers did the ERM-related papers replace? What topics related to our tried-and-true areas of expertise have been omitted from the education of new actuaries? Although we hold some monopoly power when it comes to statutory opinions, we don’t enjoy demand as inelastic as that of doctors or lawyers, nor does the CAS have as much legislative influence as the American Medical or BAR associations. While we’re plotting our expansion into non-actuarial territory, who is planning to invade our traditional areas?

As a profession, we’re taking a risk by moving into new areas, but as the Society of Actuaries likes to remind us: risk is opportunity. Without risk, there is no reward (or at least not much of a reward.) With the exam process doing a possibly too good job of restricting supply, maybe now is actually the perfect time to take a demand-side gamble. Our risk expertise should enable us to sort out the good gambles from the bad, after all. The opportunity cost, however, cannot be ignored, especially with the SOA ready and waiting to gobble us up. The time we spend sharpening our non-actuarial tools is time we don’t spend developing our traditional skills.

Speaking of opportunity cost, we could all take our membership dues and buy Facebook shares instead. Just a thought.

*I’m skeptical of how much this can help Facebook when its users are not the ones paying money. Unless a user is active on Facebook, it’s likely that Google has much better data on that person than Facebook does.

No comments:

Post a Comment