Why You Are Thinking About Income Distribution Wrong
Are There too few Millennial & Zoomer Millionaires and are the rich getting richer?
We have always had an interest in income distributions and demographics. In general it is interesting, but specifically to you, seeing how the different stats work is very motivational.
During school, our coursework and post-school exams had a heavy focus on statistics which meant we saw how a set of data can lead to widely different views. Two people could look at the same data and based on their biases, how they sorted it, and how they set relationships, they could support completely contrary conclusions.
Then you add into the mix that many topics raise lots of emotions and that most people have strongly held convictions or beliefs, and you can get to some pretty outlandish results with everyone claiming they are following the numbers.
Or as Mark Twain put it, “There are 3 kinds of lies: lies, damned lies, and statistics”
There are 3 kinds of lies: Lies, Damned Lies, and Statistics
One of the commonly used examples to back in our school days looked at the number of police vs the number of crimes in an area. This usually was used right after discussing dependent and independent variables. The dependent variable is the one that is supposed to change based on how the independent variable changes.
For example, if you looked at sales number per month. The months are independent as they happen regardless of how your sales go. Time marches on and all. But in sales, there is seasonality (think, selling pool supplies in New England, your sales are likely higher in the summer months when people actually have their pools open). So you would put months on the horizontal X axis as the independent variable and your sales on the vertical Y-axis as the dependent variable.
Cool, so now look at the chart:
If you read this chart as presented, it shows that as the number of police go up, the number of crimes go up. More police leads to more crime you could conclude.
Most of us would say “well isn’t is more likely that neither of these is independent?”. The more crime in an area, the more police there likely are. But maybe over the longer-term as the higher number of police need to justify their budgets, they start arresting more people for petty stuff leading to more broken homes and people with criminal records who can’t get jobs which then forces them into crime.
How quickly you make the jump to either conclusion based on the data likely depends on your worldview and feeling towards the currently very emotional and politically charged issue of policing. But just flipping the X & Y axis you get 2 different stories (more crime leads to more police vs more police lead to more crime).
Or you can go right to the experts and see what they have to say…
Because rich white ice cream makers in Vermont are the experts I guess??? Seriously, this was the top google result.
Anyway, we aren’t here to talk about policing or overpriced ice cream. But there are a lot of financial stats that seem to say one thing, that a lot of people have a vested interest in interpreting the data one particular way, and that get thrown around a lot.
Let’s look at 2 common ones and breakdown ho much of the headline is those lying statistics. Or as we like to say it…you can’t hiiiidddddeee those lyin stats…
For example…