Science is not the process of finding out how the world works. Science is the process of testing our assumptions and claims about how we think the world works.
The end result is indeed that we find out more, but science itself is the process, that scientific method. Anyone can make any claim they like, anyone else can contest that claim.
It is through that process that we zero in upon claims that are actually true. Or, in the way science really works, we'll accept them as being true until someone proves different.
Economics has its pretensions to being a science. So, testing the claims that people make is how we get closer to that truth we're all trying to uncover.
It is generally accepted these days that income and wealth inequality are increasing. Thus we've all got to do something to stop this happening.
Now, my own contention would be that even if this were true -- that both inequalities were increasing -- we still don't need to do anything. But, for neither side of that argument is it a scientific claim.
Some people have moral concerns about inequality, others don't. The claim that inequality is important to reduce is one of those moral claims, as is my own that it just doesn't matter. Morals are not, of course, science.
Where we get to science is in trying to analyze whether inequality is, in fact, increasing. The French trios, Thomas Piketty, Emmanuel Saez, and Gabriel Zucman have been telling us for years now that both income and wealth inequality are increasing.
Of course they're also the people who make that moral claim that this is bad and something must be done about it. But that moral claim does depend upon the scientific one, that those inequalities are increasing.
Which, well, might not be true. As a new paper analysing their work shows. When we try to replicate their results -- repeat their sums that is -- we find that they've made some assumptions along the way. And those little assumptions nearly always have the effect of increasing recorded inequality. As if, you know, they were weighting their numbers to show that inequality was increasing.
If we were very unkind about this, we might assume that they started with the desire to reduce inequality, then set out to prove that it was rising to justify that.
The thing is that every other measure of inequality that we've got shows that it is in fact reducing. The poorer countries of the world are growing faster than the richer ones. So, the poorer of the world are getting richer faster than the richer are -- that's a reduction in inequality. All of the traditional measures -- don't worry about what they each are -- like the Gini, the Palma, and so on show falls, not rises.
If we start to use non-traditional methods, then it becomes ever more obvious that inequality is falling. Things like WhatsApp, Google, or Facebook are free to use and so they don't appear in any estimates of income, wealth, GDP or, therefore, inequality. But we all have equal access at exactly the same price. That's a fall in inequality.
It's the Piketty, Saez, Zucman estimates that are out of line with all the other evidence we've got. Therefore it's those new claims that have to have substantial evidence to overturn the previous scientific result. Which, given this bias in the way they've done their numbers, might well not be true.
The challenging of these new claims is the scientific method. And science itself -- yes, even in economics -- is the set of results that have survived such challenges well enough that we consider them to be true. Until new and other challenges overturn that result, of course.
The result of all of this, at this stage, is almost certainly that inequality has not been getting worse. Rather, it has been decreasing as the traditional measurements tell us. The implication of that is that we don't have to have new policies to reduce inequality because what we're doing right now is already reducing it.
The larger lesson from this? People will make scientific and economic claims to bolster the policies they want to enact anyway. But we should insist that they're actually right and can actually prove it before we do as they ask.
Tim Worstall is senior fellow at the Adam Smith Institute in London.