I created this map based on data from Bloomberg/PayScale. For more info on what is displayed click on the i-button (3rd to right) in the top menu or ask me here.
If I'm interpreting this correctly, this means a median coding salary in USA/Canada/EU is about/slightly below average for the economy, but in the developing world (particularly India) its waaay above average?
This kind of makes sense, although it sure shatters the illusion that software jobs are well paid in the developed world.
In most countries with data, including the US, Canada and the majority of European countries, software engineers earn more than the average person. In India and Pakistan the difference is particularly striking.
> software jobs are well paid in the developed world.
The data is deceptive. In a developing country, earning the average income represents a pretty awful quality of life.
I live in the 3rd highest country on that list, earning significantly more than the median pay stated, and I'm actively attempting to emigrate to secure a better quality of life.
I'd much rather see a table that shows Software Engineer Quality of Life vs. Country. (Ideally one that includes range and percentile figures, because median is pretty useless.)
How would you measure quality of life? This income data is hopefully based on hard facts and not some kind of perception of the world.
The median income is often closer to what most people actually earn than the mean for example. I'm pretty sure the data people at Bloomberg had good reasons to chose these figures.
> How would you measure quality of life? This income data is hopefully based on hard facts and not some kind of perception of the world.
Life is all about perception. Everybody commenting here is describing their perception of these numbers.
However, the data really is deceptive. You cannot draw any conclusions whatsoever from this data alone, unfortunately. What conclusions are being drawn that are valid are actually drawing upon extra information about the countries concerned. For example:
> The median income is often closer to what most people actually earn than the mean for example
It's not really important what "most people earn". Take South Africa, 3rd on the list in this chart. This is a country with a 25.2% unemployment rate and significant income inequality [1]. The median income listed is $6,600. If you're interested in technology - good luck programming Casio digital watches, because that's all you're going to be able to afford. Even when do you save up to afford things, you die sooner than the rest of the world so you don't even get to enjoy it for long [2].
What's important is what kind of quality of life you want, and whether or not the income you earn in your chosen profession can provide for that. Over and above that are cross-cutting concerns, for instance, do you really want to live in a country with a higher homicide rate than Colombia and significant corruption? [1]
> I'm pretty sure the data people at Bloomberg had good reasons to chose these figures.
Bloomberg is providing raw data; they provide tons of other data [2] but they expect you to read them all together in a manner that make sense to you. They do not impose an opinion about how to interpret the data. what I'm saying is it would be great if there were a way to factor all the data in together in a way that allows people to specify preferences (e.g. homicide rate < 10) to determine a personalised quality of life score.
You're certainly right that looking at a single factor like median income doesn't say anything about whether I'd want to live and work in a country or not. There is more to it, but it is largely determined by my personal preference, which differs from those of other people.
I think it is extremely hard to come up with a measurement for quality of life, but there are attempts to do it, e. g. the OECD Better Life Index http://www.oecdbetterlifeindex.org/
But then too many factors can be overwhelming and hard to visualize. While the flower like shapes in the OECD example look nice, I find it hard to draw any conclusions from them.
that part isn't particularly deceptive - I expected software to be a prestigious job in developing nations, one that lifts people out of poverty.
The part I was surprised about was where the median code monkey in the developed world sat relative to the mean earnings - which was, in fact, me misreading the data (see OP's reply, a sister to your comment).
as this comment's sister mentions, median is actually much much better than mean when we're looking at pay, where a small number of high paid jobs skew results dramatically.
You are forgetting about profits which in the industrialized world is about 35% of the surplus, only 65% is wages. Therefore an average salary would only net a ratio of 0.65 in salary to gdp/capita. In other words the 0.9-1.1 ratios in the EU are pretty good.
E.g the ratio in Sweden is 1.03 and if an average salary is 0.65, it means that the average developer salary is 58% higher than the average salary. It matches my own anecdotal data pretty well.
It would have been better to compare the median developer wage to the normal median wage. But median wages can be hard to get (often you get bullshit like household median wages). Anyway, in the US median wage is a lot lower than average, so the comparison does not really say much there. In other countries the gap is often smaller, but still.
"The highest median income in absolute terms is earned by software engineers
in Switzerland (ratio 1.21) followed by Norway (81,400 USD) and the
United States ($76,000 USD). Norway has the highest GDP per capita (99,574 USD)
in this list with software engineers median income at about 82% of it."
Due to oil incomes and other stuff, Norway tends to be a serious outlier in many cases when it comes to economic metrics.
It's kind of unfair to compare annual earnings of programmers directly to GDP per capita. If you actually compare it with the annual earnings of all sectors, the ratio for Norway is close to 1.3.
Norway has very high GDP/Cap , the fourth highest in the world. A Norwegian company could probably outsource software development to the USA as a cost saving exercise. My guess is that most Norwegian software developers (that make good money) are employed in the oil industry.
Software engineer covers a huge range of roles from low level java bot to CTO. You can't just blob it together like this and expect meaningful insight.
25 comments
[ 2.8 ms ] story [ 73.5 ms ] threadThis kind of makes sense, although it sure shatters the illusion that software jobs are well paid in the developed world.
The data is deceptive. In a developing country, earning the average income represents a pretty awful quality of life.
I live in the 3rd highest country on that list, earning significantly more than the median pay stated, and I'm actively attempting to emigrate to secure a better quality of life.
I'd much rather see a table that shows Software Engineer Quality of Life vs. Country. (Ideally one that includes range and percentile figures, because median is pretty useless.)
The median income is often closer to what most people actually earn than the mean for example. I'm pretty sure the data people at Bloomberg had good reasons to chose these figures.
Life is all about perception. Everybody commenting here is describing their perception of these numbers.
However, the data really is deceptive. You cannot draw any conclusions whatsoever from this data alone, unfortunately. What conclusions are being drawn that are valid are actually drawing upon extra information about the countries concerned. For example:
> The median income is often closer to what most people actually earn than the mean for example
It's not really important what "most people earn". Take South Africa, 3rd on the list in this chart. This is a country with a 25.2% unemployment rate and significant income inequality [1]. The median income listed is $6,600. If you're interested in technology - good luck programming Casio digital watches, because that's all you're going to be able to afford. Even when do you save up to afford things, you die sooner than the rest of the world so you don't even get to enjoy it for long [2].
What's important is what kind of quality of life you want, and whether or not the income you earn in your chosen profession can provide for that. Over and above that are cross-cutting concerns, for instance, do you really want to live in a country with a higher homicide rate than Colombia and significant corruption? [1]
> I'm pretty sure the data people at Bloomberg had good reasons to chose these figures.
Bloomberg is providing raw data; they provide tons of other data [2] but they expect you to read them all together in a manner that make sense to you. They do not impose an opinion about how to interpret the data. what I'm saying is it would be great if there were a way to factor all the data in together in a way that allows people to specify preferences (e.g. homicide rate < 10) to determine a personalised quality of life score.
[1] http://www.bloomberg.com/visual-data/best-and-worst/most-str...
[2] http://www.bloomberg.com/visual-data/best-and-worst/longest-...
[3] http://www.bloomberg.com/visual-data/best-and-worst/
I think it is extremely hard to come up with a measurement for quality of life, but there are attempts to do it, e. g. the OECD Better Life Index http://www.oecdbetterlifeindex.org/
But then too many factors can be overwhelming and hard to visualize. While the flower like shapes in the OECD example look nice, I find it hard to draw any conclusions from them.
The part I was surprised about was where the median code monkey in the developed world sat relative to the mean earnings - which was, in fact, me misreading the data (see OP's reply, a sister to your comment).
as this comment's sister mentions, median is actually much much better than mean when we're looking at pay, where a small number of high paid jobs skew results dramatically.
E.g the ratio in Sweden is 1.03 and if an average salary is 0.65, it means that the average developer salary is 58% higher than the average salary. It matches my own anecdotal data pretty well.
It's kind of unfair to compare annual earnings of programmers directly to GDP per capita. If you actually compare it with the annual earnings of all sectors, the ratio for Norway is close to 1.3.