Behavioral Economics, Insurance and Making Healthy Choices

February 14, 2011 at 1:02 pm Leave a comment

I was reading this piece on Dark Daily, entitled “Behavioral Economics Likely to Push Up Utilization of Clinical Pathology Laboratory Tests” which suggests that laboratory usage will increase due to a trend in insurance companies to lower premiums through proactive intervening tests instead of costly reactive procedures. These tests would measure and inform certain healthy/unhealthy behaviors and would influence the price of coverage for individuals.

The piece’s author, Michael McBride, believes that more people will choose for less expensive coverage in exchange for living healthy, which will result in more testing sent to clinical and pathological labs.

While I thought it was interesting – especially as we at mTuitive have a pathology product – I was unsure about the validity of McBride’s assumptions. Luckily, I’m fairly familiar with behavioral economics as my friend Ryan has been studying it for years and is currently in the economics PhD program at Duke University. I asked Ryan if he could clarify for everyone the definition of behavioral economics and provide any examples that either support or contradict McBride’s findings. Here is Ryan’s response:

Behavioral economics is a burgeoning field due to its robust nature in explaining economic decisions. Where many view standard economics as too rigid, relying on assumptions of consistent preferences, full information, and unbounded rationality, behavioral economics use of flexible concepts like social/cultural framing, the status quo, and loss aversion seem to be more representative of the “real” world. Due to its accommodating assumptions and straightforward approach, though, there is a tendency for people to simplify or generalize the predictions of behavioral economics. An example of this trend can be seen in the Dark Daily article, “Behavioral Economics Likely to Push Up Utilization of Clinical Pathology Laboratory Tests.”

The article presents a well thought-out premise; new insurance schemes which incentivize improving health will have large take-up and thus subsequently lead to major increases in clinical lab tests. To motivate this discussion the author states, “given a choice of either unhealthy activities (e.g., smoking, eating badly, not exercising) coupled with an expensive health benefit plan, or an inexpensive, even zero cost, health plan that promotes healthy choices, behavioral economics theory predicts that consumers eventually choose the latter. That choice should lead to improved health while driving down the cost of healthcare.” This statement though is not completely valid. Nothing inherent in “behavioral economics”, or standard neo-classical economics for that matter, makes a costly insurance program with no behavior related stipulations necessarily the preferred choice over a cost-less but regulated alternative.

Predicting how people would react to these different insurance instruments is completely reliant on how one models the preferences of the consumer. If being in the regulated insurance program brings the consumer dis-utility from following the stipulations of the “healthy choices” plan or if current day enjoyment of behaviors (unhealthy choices) trumps long-term happiness derived by the expectation of future health (i.e. if the consumer is impatient enough) standard economics would predict that the costly insurance would continue to dominate the market. Moreover, when the findings of behavioral economics are used to predict the insurance choice, the outcome is increasingly unclear.

One prediction of behavior economics is that the way we contextualize particular decisions matter. Simply being offered two goods can be a very different choice problem depending on the circumstances surrounding the decision. In the case of insurance and health behaviors this context factor may be of great importance.

For example, it has been shown that when one views a decision as a market transaction versus social construct behaviors can vary greatly. Experimentally it has been shown that offering a wage for an activity that had been done freely before, and previously had been thought of as a “social duty”, actually can decrease participation in that activity. According to standard economics this should never happen, but when one is offered money for an action, the activity now becomes framed in terms of the market, and if that wage is not as high as they think they deserve, a previously freely done good deed can be interpreted as a poorly paid “job”. This contextualization issues applies in this insurance decision as well. If making healthy decisions is a choice motivated because it is “the right thing to do” or because it is more “socially acceptable” putting a price on acting unhealthy may legitimize for some the trade-off between paying for health insurance and acting unhealthy (encouraging “moral hazard” behavior). Additionally, behavioral economics is very keen to pursue how people deal with risk and loss. Common themes of this field of study are loss aversion, risk aversion, and ambiguity aversion, and each has its place in the insurance decisions.

Many studies have shown that people would prefer to take a less advantageous position in which there is not risk for failure. Not being able to succeed can be very disheartening and thus motivate risk-insulated behavior. In this way sticking to the “status quo” plan and avoiding the insurance in which one’s behavior is evaluated can be the optimal choice. Further, if people fear that failure to achieve levels of healthy behaviors will cause them to be dropped by the insurance plan or can drastically shift their budget, people may prefer the more consistent and non-action dependent plan than one which contains some ambiguity related to evaluations which the consumer fears they can not control (fear of failure to self commit), feel has some amount of chance (risk-aversion), or feels they don’t fully understand (ambiguity aversion).

Predicting economic behavior is of utmost importance to those who wish to create policies that incentivize certain outcomes. While standard economics provides a theoretical framework to do this, its assumptions can feel too constricting to interested parties. The field of Behavioral economics has opened the door to more flexible models of human decision-making and its use is imperative to proper development of policy prescriptions. With its flexibility though comes complexity and in most cases behavior economics, at a baseline level, cannot give strict predictions of outcomes, as with all models, the results will be conditional on the assumptions of the researcher. Thus, when making predictions about the impact of a policy it is important to combine the intuitive modeling of behavioral economics with rigorous empirical investigation and then one can start to make informed market and behavioral analysis of the introduction of a new good.

So what’s the take away? In the end, as William Goldman says, “Nobody knows anything.” Well, that’s a bit too cynical. But it’s too early to tell if, as these behavior and lifestyle dependent insurance programs are rolled out, they will be widely embraced.

Sure, if people know that they will pay less if they follow certain criteria or benchmarks that they should already be hitting, (like smoking cessation, etc.) that may encourage them to adopt the healthier lifestyle and therefore utilize more clinical testing. At least, that makes a certain amount of cognitive, logical sense.

But behavioral economics isn’t the study of logical actions – it’s the study of actual actions. Where people can be swayed by fear, or anxiety, or some form of rationalization that allows them to circumvent the logical. For those who are worried about being dropped from coverage for not making weight, or those who simply look at the extra cost for the “regular” insurance as a “fat tax” – then they have no problem with the current reactive system. Which would then translate to no change to lab tests.

It’s possible that this trend of applying “behavioral economics” could translate into more money for labs. But, by ascribing any probability to it in its infancy, the article’s conclusion is an assumption that is based more on McBride’s (hopeful) bias than on any previously documented behavior.


Entry filed under: General. Tags: , , , , , , , , , , , , , , , , , , , , , , , , .

Best Behavioral Study Ever? Movin’ Out!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

Trackback this post  |  Subscribe to the comments via RSS Feed

Wholly Owned Subsidiary of mTuitive


Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Join 3 other followers

mTuitive on Twitter!


Disclosure Statement - The authors of this blog are paid employees of mTuitive Inc. and are compensated for their services.

%d bloggers like this: