Rankings of average wages by country are built from a mean rather than a median, converted at either the market exchange rate or a purchasing power factor, measured either before or after tax, and covering either all employees or only full-time ones. Each of those four choices moves the order, sometimes substantially. A ranking is therefore an answer to a specific question, and the question is almost never printed alongside the table.
Why two credible rankings disagree
Search for the countries with the highest average wages and you will find several lists that do not agree with each other, published by organizations with no obvious reason to be wrong. It is tempting to conclude that one of them is using bad data. Usually none of them is: they have made different methodological choices, each defensible, and the choices are doing the work. Reading a ranking without knowing which choices were made is a bit like reading a race result without knowing the distance.
There are four such choices, and once you know what they are you can look at any ranking and work out roughly why it says what it says. That is a more useful skill than memorizing the top ten, because the top ten changes and the four choices do not.
Choice one: the mean or the median
Nearly every published country ranking uses the mean, because it is what the underlying national accounts produce most readily. The mean is pulled upward by high earners, and it is pulled upward by different amounts in different countries because the shape of the distribution differs. A country with a large concentration of very high incomes will post a strong average even where the typical worker earns considerably less than the number suggests.
The median describes the person in the middle and is far more informative about what a normal job pays. It is published less often, and where you can find it, it is the figure to prefer. When a ranking gives you only a mean, the honest reading is that it tells you about the country’s total wage bill divided by its workers, which is a real fact about an economy and a poor guide to your own prospects there.
Choice two: the conversion
Converting every country’s wages into one currency requires a rate, and the two available rates answer different questions. The market exchange rate tells you what a salary is worth if you moved it abroad. A purchasing power factor tells you what it buys at home. Rankings built on the market rate flatter countries with strong currencies; rankings built on purchasing power flatter countries where local prices are low.
Neither is wrong and they produce noticeably different orders. If you are reading a ranking to decide where you might live and work, the purchasing power version is the one that answers your question. If you are reading it to understand global capital flows, the market rate version is. Check which one is on the page before you take the order seriously.
Choice three: gross, net, or something in between
Most rankings quote gross pay, which means the country with the heaviest deductions and the country with the lightest are being compared on a figure that neither worker ever sees. A ranking on net pay reorders things considerably, generally moving low-tax economies up and high-contribution ones down.
The complication is that net pay comparisons then understate the high-contribution countries in a different way, because the deductions are buying healthcare and pensions that the low-tax worker has to fund separately. There is no single figure that resolves this cleanly, which is precisely why several exist. Some publications split the difference by reporting net pay for a specified household type — a single person on the average wage, or a couple with two children — which is more honest and immediately raises the question of whether that household is anything like yours.
Choice four: which workers are counted
Some series cover all employees; others cover only full-time employees, or convert part-timers into full-time equivalents. Where part-time work is common, including it drags the average down, and countries with high part-time employment therefore look weaker on an all-employees basis than on a full-time one.
This is not a small adjustment in the countries where it applies, and it is the choice most often left unstated in the write-up. Self-employment is a related question — usually excluded, which matters more in some economies than others. If a ranking’s methodology note is one sentence long, this is the detail it left out.
What a country-level figure cannot tell you
Even a perfectly constructed ranking is answering a question about a country, and you do not have a job in a country. You have one in an occupation, in a city, at a level. National averages conceal spreads within a country that are frequently wider than the differences between countries, and the occupation you work in can matter far more than the border you are standing on.
A software developer and a retail supervisor moving between the same two countries will experience completely different changes in pay, because their occupations are priced by different labor markets that happen to sit inside the same national statistics. The country ranking averages over both of them and tells you about neither.
Where the rankings are genuinely useful
They are good for the coarse question. If you are trying to work out which handful of countries are worth investigating at all, a ranking gives you a defensible shortlist in a few minutes, and that is a real service. They are also useful for direction of travel over time, since a country’s position moving steadily up or down across a decade is telling you something durable.
They are poor for the fine question, which is the one you actually have. Once you have a shortlist, drop the rankings entirely and go looking for occupation-level data in each candidate country, or for advertised ranges in current job postings. The ranking got you to the shortlist; it cannot get you any further, and treating it as though it can is where the misreading starts.
How to read one in thirty seconds
Find the methodology note, which is usually a line under the table or a link at the bottom. Establish whether it is mean or median, market rate or purchasing power, gross or net, and all employees or full-time only. If it answers all four, you can trust the ordering for what it is.
If it answers none of them, the table is decoration. That is not a reason to be cynical about every ranking, because plenty of them do document their choices properly. It is a reason to spend thirty seconds checking before you let a table shape a decision about where to spend the next five years.
Common questions
Why do country wage rankings disagree?
Because each makes four methodological choices — mean or median, market rate or purchasing power, gross or net, and which workers are counted. Each choice moves the order, and they are rarely stated.
Should I prefer the mean or the median?
The median, wherever you can find it. The mean is pulled up by high earners, by different amounts in different countries, so it describes the wage bill rather than a normal job.
Which conversion rate should a ranking use?
Purchasing power if you are asking where to live and work; the market rate if you are asking about global flows. They produce noticeably different orders.
Is gross or net pay the better basis?
Neither resolves it cleanly. Gross compares figures nobody receives; net understates countries whose deductions buy healthcare and pensions the low-tax worker funds separately.
Why does it matter which workers are counted?
Where part-time work is common, including it drags the average down. Countries with high part-time employment look weaker on an all-employees basis than a full-time one — and this is the detail most often omitted.
What can a country ranking not tell me?
Anything about your job. Spreads within a country are frequently wider than differences between countries, and two occupations moving between the same pair will experience opposite changes.
So are rankings useless?
No — they are good for building a shortlist in a few minutes, and for direction of travel over a decade. They are poor for the fine question, which is the one you actually have.
How do I check a ranking quickly?
Find the methodology note and establish the four choices. If it answers all four, trust the ordering for what it is. If it answers none, the table is decoration.