Labour economists devote much of their time to looking at what factors affect wages. The classic regressions they run show that you are likely to have a higher wage if you’ve been to school for more years and if you have more work experience. This seems pretty intuitive and obvious.

They’ve also looked at how IQ/intelligence affects wages. The general consensus is the common sense one – you earn more if you’re smarter. Again, this makes sense intuitively. There’s a nice paper written by Zagorsky (2007) that goes through this (and more) in quite a bit of detail if you’re interested.

However, from my brief time searching the net for more information on the matter, it seems as though most people have been looking for a linear relationship.

My theory

I think the relationship between intelligence and wages is definitely non-linear. Based purely on my own intuition here’s what I think the relationship is:

The graph above shows basically how I feel wages and intelligence are related to each other (Wages are represented on the vertical axis and IQ on the horizontal). The relationship is positively skewed – here’s why:

  • If you are below average intelligence, then clearly your earnings will be relatively low. At this point, when intelligence increases, so will your wage. This is because at that level of intelligence, if you became a few points smarter, it might (for example) mean the difference between being able to use a machine and not.
  • There will be a point (somewhere a bit above average) where salary will be maximised. In other words, I believe that there is an optimal IQ level that results in you earning the most money. At this level, you are smart enough to do the various tasks required in a ‘high-skilled’ job. However, you are not smart enough to be a scientist or an academic. I think this is good for earnings because once you cross that threshold of intelligence, your intellectual curiosity coupled with cynicism for the social infrastructure around you makes it difficult to ‘settle’ into a wage maximising lifestyle.
  • When you are extremely smart (IQ well above average), those high paying finance and business jobs may not exactly fulfil your intellectual curiosity. So, you may settle into something orientated around research, academia, science etc. These jobs require a high level of intelligence, but aren’t as easy to ‘get rich’ from (though I appreciate it’s not impossible). Hence, the final part of the graph has wages tailing off, but obviously not at the same rate as the below average IQ group. A researcher will still expect to earn more than a cleaner.

There’s also a case for there to be another bump in the graph at very high intelligence levels – the Einsteins of the world (if they get noticed during their lifetime!) may come up with something so brilliant that they do indeed become very rich. But I think this only corresponds to a few extreme cases.

Finally, I also think we can observe this phenomenon in some single occupations. For example, a banker may be satisfied making a lot of money from their daily investing. These are relatively intelligent people. But an extremely intellectual banker may find that he/she is more interested in the mathematics of finance. Hence, he/she chooses to delve into the theory by becoming an academic rather than working for an investment bank each day. This has the cost of a reduced income.

Closing remarks

It’d be nice to get some good data on this to test this hypothesis a bit more scientifically. If you know of any IQ inclusive labour market datasets, please let me know! I did briefly look at the National Longitudinal Survey of Youth – the popular US dataset that Zagorsky uses in his paper, and that many economists have used extensively. In this, the relationship between intelligence (measured via the Armed Services Vocational Aptitude Battery or ASVAB) and earnings is clearly positive, but it’s quite tricky to come up with any conclusion more comprehensive than that without devoting a lot of time to it.

Maybe my theory applies only to males, or certain groups of people. But I’m fairly convinced that this sort of pattern exists.

If you have any thoughts on this, or have done research yourself, I’m more than happy to discuss them!