Disclosure policies often do change people’s behavior, but not always in the ways or for the reasons that the policies’ designers intended. (More)

Psychology and Policy, Part II: Disclosure, in Practice

This week Morning Feature considers how insights from psychology and behavioral economics should shape public policy. Yesterday we began with the standard economic theory of disclosure, which libertarians and conservatives often cite as the solution for all market failures. Today we see how research in psychology and behavioral economics challenges the standard theory, and how that research can improve public policy on disclosure. Tomorrow we’ll conclude with how insights from psychology could improve our discussion of climate change.

“Yeah, whatever”

You need a left-handed widget wrench and what better place to go than BidaWidget? They have great prices, and they have that top of the line widget wrench, with the chrome finish and the hollow handle where you can keep extra left-handed widgets. You click “Open an Account” and enter your email address and your mailing address and your card information and click “Buy Widget Stuff Now!” … but that button is gray and nothing happens. Oh, duh. First you have to click “Accept Our Privacy Policy,” the button next to “Read Our Privacy Policy.”

“Yeah, whatever,” you mutter. You don’t have time to read all that legalese. You want to order that widget wrench, the really good one with the chrome finish and that nifty holder for extra left-handed-widgets. Besides, BidaWidget has a privacy policy and that means your personal information is safe, right?

Congratulations. You’re like the 97% of us who don’t read online privacy policies, and the 75% who think a privacy policy means the site protects customers’ privacy. Oops.

Several states including California require online marketers to disclose how they use customers’ data and those long passages of legalese declare how much privacy you give up when you click “Accept Our Privacy Policy.”

Standard economic theory says you’ll read that disclosure, weigh your concerns for privacy against the low prices and other goodies that BidaWidget offers (free next-day shipping on orders over $20!) and make a rational choice. In theory, the difference between the low price at BidaWidget and the higher price at a source that really does protect customers’ personal invormation is the market price for privacy. In theory, if enough widget customers value their privacy highly enough, they’ll shop elsewhere and BidaWidget will have to improve their privacy policy, lower their prices even more, or offer more goodies (24-hour customer support!).

“Limited attention”

In practice, only 3% of online shoppers read privacy policies … because you’re doing that project but you only have a right-handed widget wrench (you can never do a project without buying at least one new tool!) and you don’t want to wade through a bunch of legal mumbo-jumbo. You want to order a left-handed widget wrench (the new ones are ergonomic designed with the pebbled-rubber coating on the handle!) so you can get back to work on your project tomorrow (because free next-day shipping!).

In short, you have limited attention, as Loewenstein et.al. explain:

There are serious limitations on the amount of information to which people can attend at any point in time. The standard economic account would emphasize that attention is a scarce resource and suggest that people make rational (even if fairly rapid) decisions about how to allocate it. Research in psychology, by contrast, suggests that people do not decide how to allocate attention; certain items capture attention while others disappear into the background, even if they are exceedingly important, and even if it would be rational to focus on them.[…]

Disclosures are so ubiquitous that we tend to be unaware of them, and, when the implicit is made explicit, one cannot help but be struck by the impossibility that anyone could attend to even a fraction of the disclosures to which we are exposed. [Citations omitted.]

“Consumers should ideally draw the logical inference….”

In theory, we’ll also notice what isn’t disclosed and be suspicious because it was left out. In practice … not so much:

We have already discussed research showing that people have only limited capacity to attend to information that they are presented with; other research shows that people typically pay even less attention to the absence of information than to its presence, even when both are equally informative.

More evidence of inattention to missing information in a real world market context comes from research examining the cold-release of movies – i.e., the release to consumers without first giving access to reviewers. Studios cold-release movies when they are confident that the reviews will be unfavorable, and consumers should ideally draw the logical inference from the release of movies with no prior reviewer coverage. However, Brown, Camerer, and Lovallo find that, in fact, cold-released movies do initially better than movies that are pre-released to critics only to receive predominantly negative reviews. [Citations omitted.]

We’re also prone to motivated attention. Studies show people log on to check their online trading accounts more often when the market goes up, and avoid them when the market goes down. Other research found that people most at risk for HIV are the least likely to get tested, because they fear getting a diagnosis. Smokers ignore warning labels on cigarette packs and are more likely to hit the mute button or change channels when gruesome anti-smoking ads come on TV.

And even when we do pay attention to disclosures, we may not do so for the reasons the policymakers had in mind. For example, many states require restaurants to disclose the calorie count and nutritional information for items on their menus. In theory, people who read the disclosures will choose more healthful foods. In practice, poor people who read the disclosures are more likely to choose menu items that pack in the most calories for the lowest price … very rational, but not what the menu-disclosure policy makers had in mind.

“All else held equal, people prefer to tell the truth”

Disclosures can also have surprising effects on the person who reveals the information, as Loewenstein et.al. explain:

There is by now a very large literature in behavioral and experimental economics demonstrating what people outside the profession might find obvious – that people are powerfully driven by other-regarding motivations such as altruism, fairness, and a desire to perceive themselves as good people, and that, all else held equal, people prefer to tell the truth and also expect others to do so. [Citations omitted.]

Let’s say your doctor gets a referral fee for patients who enter a clinical trial. She’s a good doctor and she wants to be a good person, so she’s careful to refer patients only if she has a good reason to think the trial would help them. Until, that is, she’s required to tell patients about that referral fee:

The fact that people are intrinsically motivated to provide unbiased advice and high quality products (even when they could pass off inferior ones to naïve consumers) is important, because disclosures of conflicts of interest can, in some cases, undermine such motivation, a phenomenon that Cain, Moore and Loewenstein dubbed ‘moral licensing.’ Moral licensing occurs when the perception that an advisee has been warned, via disclosure, of an advisor’s potential bias makes the advisor feel less responsible for giving unbiased advice. [Citations omitted.]

Moreover, studies show that people are more likely to accept bad advice if the advisor has disclosed a conflict-of-interest. The patient may feel the doctor is subtly asking the patient to help her get that referral fee (the panhandling effect), or that asking for time to get a second opinion about that clinical trial will be taken as a sign of distrust (insinuation anxiety).

Thump-ump, thump-ump

On the other hand, merely knowing we must disclose information will often make us more ethical, even if most people ignore what we disclose:

In a case often cited as a paradigm for successful disclosure, for example, hygiene ratings of restaurants in L.A. affected patronage patterns, which then motivated restaurants to improve their sanitation practices. But in many situations (to some extent including the case of restaurant hygiene ratings), an industry response can be found amidst little evidence of a consumer response.

This pattern raises an obvious question: Why are providers changing their products in response to disclosures that their customers are largely ignoring? On the basis of profit considerations alone, consumer inattention should lead producers to do exactly what they were doing before. Evidently some disclosers either have an exaggerated expectation of the likely consumer response or feel guilty about the information disclosed. We suspect that sellers may well have an inflated sense of the public salience of disclosures, in a phenomenon related to the spotlight effect, by which people exaggerate how much other people are looking at them, and also analogous to the confession of the protagonist in Edgar Allen Poe’s famous short story, The Telltale Heart, who imagines that the police can hear the heartbeat of the man he has killed and buried beneath the floorboards of his apartment.

Making disclosure work

The authors offer several suggestions to make disclosure policies more effective:

  • Simplification – If people don’t read or can’t understand a disclosure, its only effects will be those on the discloser (such as moral licensing, or the spotlight effect). The authors suggest changing disclosure mandates to reduce the number of less-important items, so people will recognize the more-important items.
  • Standardized and comparative information – People make better choices when disclosures enable them to compare similar products or services and balance relevant tradeoffs. As positive examples, they cite energy efficiency ratings for appliances and automobiles.
  • Social comparison information – We can encourage good behavior simply by publishing it, such as Forbes list of top philanthropists, or sending out ‘energy report cards’ that score customers’ power bills against their neighbors’. But we must be aware of possible counter-effects: in one study, Republicans who had top energy report cards – used less power than their neighbors – increased their usage after they saw their scores.
  • Vividness – Disclosures with vivid images are more effective than mere words … unless they turn away the very people who should pay attention (such as smokers muting or changing channels to avoid those gruesome TV ads).
  • Smart disclosures and intermediaries – Disclosures may be more effective if delivered through a neutral intermediary, to reduce the moral licensing, panhandling effect, and insinuation anxiety responses.

Finally, and perhaps most important, policymakers must recognize the limits of mandatory disclosure policies:

As we have shown, disclosure holds considerable promise as a tool of public policy, especially as a means of altering the behavior of disclosers as opposed to disclosees. However, it also has severe limitations, and can backfire in certain situations, damaging the interests of those it is intended to help. Given these limitations, and the always present temptation of taking the path of least resistance, policy makers need to be vigilant against the risk that mandatory information disclosure policies will be implemented as a substitute for other, often more effective, regulatory interventions.

Mere disclosure will not prevent all market failures, because we are not purely rational economic actors. As the authors conclude, when it comes to the effects of disclosure … “Psychology Changes Everything.”


Happy Friday!