Posted by: structureofnews | September 4, 2012

Means And Ends

Is sunlight the best disinfectant?

It’s not a small question, given that much of the push for greater freedom of information and transparency is built around the notion that more and better disclosure helps government and democracy function better.  And it certainly makes sense – making it harder to hide illegal or immoral government or corporate activities should help cut down cases of such wrongdoing.

Enter Dan Ariely, the economist who has made a career out of testing such notions – and upending our preconceptions about the world.  His best-selling Predictably Irrational chronicled a rigorous experiment-based approached to economic theories – so-called “behavioral economics” – and showed how our ideas of human rationality were hugely flawed.  And in his latest book, The (Honest) Truth about Dishonesty, he looks at the impact of disclosure on the quality of advice people get.  It ain’t pretty.

He cites an experiment where subjects were asked to guess the amount of loose change in a jar.  The more accurate their guess, the more they made.  Pretty straightforward.  Then they were paired with an “advisor,” another subject who was given a little more information about what was in the jar – so in theory, at least, they could help the “estimators” do a better job.  In the first test, the advisor were rewarded based on how accurate the estimators were – sort of like a stock broker who makes a cut of what you make (or takes a loss on what you lose.)

But in the second test, advisors were paid more if estimators over-estimated the amount in the jar – so they had an incentive to shade their advice on the high side.  And they did: Their advice came in, on average, about 25% more than in the first test.  So far, so predictable.  But in the third scenario, the reward structure was made clear to everyone – so it was obvious to estimators that the advisors had a clear conflict of interest.  All this was transparently disclosed.  So what happened?

The advisers upped their advice even more – and the estimators, forewarned, discounted it.  But not by nearly enough.  In other words, when everything was disclosed, the advisers gave poorer advice, and the estimators did even worse than before.  That’s a rather sobering conclusion.  Ariely notes:

The main takeaway is this: Disclosure created even greater bias in advice. With disclosure the estimators made less money and the advisors made more,  Now I am not sure that disclosure will always make things worse for clients, but it is clear that disclosure and sunshine policies will not always make things better.

Ouch.  OK, so it’s just one experiment.  But if it’s broadly true, where does that leave advocates of greater transparency and disclosure?

It reminds me of the “Asian Values” argument that raged – at least in Asia – in the mid- and late 1990s, part of which was about whether democracy helped or hindered economic development.  Lee Kuan Yew, the founder of modern Singapore, fell clearly into one camp.  I covered the speech in Manila where he famously said that:

I do not believe democracy necessarily leads to development. I believe what a country needs to develop is discipline more than democracy.

And certainly there was much debate about whether nations should focus on building economic growth before developing democratic institutions, or where the latter helps the former.  Not unlike a discussion about whether transparency leads to better outcomes for citizens or not.

But perhaps it doesn’t matter if it does or it doesn’t.  As Nobel Laureate Amartya Sen noted in Development As Freedom – and I’m taking huge liberties paraphrasing – don’t we want democracy anyway, whether or not it leads to greater wealth and development?  Or as he said in a later speech:

…political freedom is a part of human freedom in general, and exercising civil and political rights is a crucial part of good lives of individuals as social beings. Political and social participation has intrinsic value for human life and well-being.

So, too, transparency.  Is that a fairly basic part of our social contract with government and many of the entities it regulates?  Does it matter if that leads to better outcomes, more wealth or less corruption?  Well, yes, it does – but an argument for greater transparency doesn’t have to rest simply on better outcomes.  The means matters as much as the ends.

And there are broader arguments in favor of more disclosure, in any case, including some from data journalism.  The more information that’s broadly available – in standard formats, as Dan Conover notes – the more opportunities to build new innovations and products in what’s clearly a much more information-led world and economy.  True, that’s an ends-based argument, and stands the danger of being overturned by facts (damn! don’t you hate facts?).

But there is an enduring human argument for better understanding of the world – and that needs more and better transparency, regardless of outcomes.


  1. Reg — As you point out, Ariely cites only a single experiment. There’s now substantial scholarly research demonstrating that media freedom — which is tied closely to transparency — is associated with corruption control, higher income and investments, and political stability. And when you combine media freedom with improvements in the rule of law, the impact on curbing corruption is particularly impressive. In other words, sunlight is indeed the best disinfectant. (Check out p.11 of Empowering Independent Media:

    • David, I don’t disagree. And I do believe there’s a good ends-based case for transparency. But I think it’s just as important to make the means-based case – that regardless of outcome, transparency is a good thing, and we shouldn’t have to depend on proving causality to make that argument. Reg

  2. I think the research makes sense: To some degree, presumably, fiduciaries as a group temper their naked self-interest when they know their clients aren’t aware of the perverse incentives, but feel freer to indulge themselves if the client knows what’s up (caveat emptor, and all that).

    But there are some key corollaries that I think may not be reflected here: First, transparency *gives you the tools to test exactly this kind of effect.* That can’t be underestimated; without knowing the bias, you couldn’t do that.

    Second, transparency may be *necessary but not *sufficient*. So in the securities context (where I’ve spent most of my time thinking about disclosure and transparency), the argument is that disclosure alone isn’t enough — you also need enforcement.

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


%d bloggers like this: