The Perils of Data Optimism

In one graph:

The world is getting better, comrade!

The world is getting better every year, they tell us. Fewer people dying in war, poverty lifting, education increasing, and most importantly, monotonic GDP growth across the board.

And yet something feels off. Various lines of argument suggest things are not as they should be, that things are in important ways teetering on the brink, and that many key intangibles are getting worse. Or at least one can find many smart people claiming it is so.

"Nonsense," the optimists say, "everything is getting better. Any doubt of the superiority and general upward trajectory of the current system can only be based on ignorance, or worse, bitterness at change."

We are fans of data. As much of it as possible to inform judgement is the ideal. But data can never be a substitute for wise judgement, and can never be conclusive on its own. Our hypotheses must always be consistent with the data, but a hypothesis based on data is no guarantee at all, especially when other, non-data arguments suggest nasty things on the horizon.

This has all been said before, 160 years ago even, by smarter men than us:

A witty statesman once said, you might prove anything by figures. ...

... Tables are abstractions, and the object a most concrete one, so difficult to read the essence of. There are innumerable circumstances; and one circumstance left out may be the vital one on which all turned. Statistics is a science which ought to be honourable, the basis of many most important sciences; but it is not to be carried on by steam, this science, any more than others are; a wise head is requisite for carrying it on. Conclusive facts are inseparable from inconclusive except by a head that already understands and knows. ...

... With what serene conclusiveness a member of some Useful-Knowledge Society stops your mouth with a figure of arithmetic! To him it seems he has extracted the elixir of the matter, on which now nothing more can be said. It is needful that you look into his sad extracted elixir; and ascertain, alas, too probably, not without a sight, that it is wash and vapidity, good only for the gutters. ...

... Twice or three times we have heard the lamentations and prophesies of a human Jeremiah, mourner for the poor, cut short by a statistic fact of the most decisive nature: How can the condition of the poor be other than good, be other than better; has not the average duration of life in England, been proved to have increased? ...

How indeed?

Perhaps there was some critical statistic besides GDP that would have obviously predicted collapse and shut the optimists up in the case of the USSR, but all accounts I've heard were that the intelligent classes didn't see it coming.

I can't help but suspect that today's data optimists would have beaten a pessimistic Russian over the head with their wonderful data until finally and catastrophically, their optimism proved unfounded.

While we should be wary of predicting something outside of the obvious trends in the data, it is good to remember that it is never certain and never the whole story, especially if we suspect by other means that something else might be going on.