More on Inequality and Misuse of Statistics

Ever taken a statistics course? Most stats lecturers devote a special moment to highlight how statistics can be misused. For instance my tutor once showed us a graph like this:

Coincidence?

We can clearly see that there’s a correlation between ice-cream consumption and deaths by drowning. But what can we infer from this? It’s possible that eating ice-cream causes drowning (due to stomach cramps while swimming). It’s also vaguely possible that drowning deaths cause increased ice-cream consumption (mourning relatives might go for an ice-cream to cheer themselves up). However the most sensible explanation is that both ice-cream consumption and drowning deaths increase is due to another factor: the weather. People eat more ice-cream and go swimming more often in summer.

However, such a straightforward explanation is hardly ever seen in economics. The empirical approach often remains unquestioned. Consider this syllogism: in the past, taxes were low. Today, taxes are high. We were poor in the past but now we are rich. Therefore, increasing taxes causes prosperity. Such a view is completely ridiculous, yet almost completely unquestioned.

Or take the suggestion in the graph below. Income inequality sharply increased before two major recessions. Therefore it’s fair to assume that one causes the other.

Why yes, yes it is.

Two possible conclusions one can reach from this graph are:

A) economic boom causes incomes in the top percentiles to rise

B) severe income inequality causes a recssion.

In my view, both are incorrect, or at least incomplete. A more accurate interpretation is:

C) years of easy money policy from a counterfeiting racket cause the economic boom. Rising incomes, and rising income inequality follow from this. This prosperity however is artificial and must eventually come to an end. Incomes, and income inequality fall during the recessionary phase.

There are however some people who argue in favour of proposition B. They are wrong.

Their theory on why income inequality goes something like this. As incomes becomes more and more unequally distributed, a dreadful scenario approaches. Since the rich save a higher proportion of their money than the poor, and since the poor are deprived of income there’s less spending taking place in the economy. The rich have the money but they don’t spend it. The poor don’t have enough money to spend. As a result, the entire economy goes into decline.

The counter to this can be summarised in a few points:

• It’s the job of entrepreneurs to successfully forecast future market conditions. The wealth inequality thesis can’t account for the cluster of errors and an economic crash. If income inequality were the main cause of recession why doesn’t the economy go into a gradual slowdown?
• At any level of income, at least some level of  consumption will continue to occur. It’s also doubtful that the rich do in fact spend proportionately less that the poor.
• Even if the rich save more than the poor, savings constitutes demand for producers’ goods. In effect, to save is to spend, and savings is the most important element of a prosperous economy.
• An economic crisis is usually marked by undersaving not underconsumption.

Thanks, Murray.

We have seen that it’s easy to turn your ideological desires into a relatively plausible science. It’s even easier if you have fancy graphs at your disposal.