Can we say, in general, what is the relationship between relative government expenditure (or the total taxation) and the average quality of life?

  • No. For one thing, quality of life is an individual and subjective matter. One person may feel that their QOL is improved by high taxation; another may feel that the same tax diminishes their QOL. – jamesqf Dec 23 '18 at 19:29
  • 2
    Also, if the government taxes 50% of income, and uses 49% of it to fund a military, that's very different from taxing 20% of income and using 19% to fund social programs. – Bobson Dec 26 '18 at 12:56
  • 3
    The same relationship as the sales of ice cream and the murder rate in New York City. – Jared Smith Dec 26 '18 at 16:48
  • Schrute Bucks::Stanley Nickles – acpilot Dec 27 '18 at 5:47

Triggered by the somewhat arbitrary selection of countries in the OECD graph in Bobson's answer, I found that data are easily available for much more countries than just the OECD countries.

Summary: there is some positive correlation between tax : GDP ratio and happiness score, but the correlation is too low to allow meaningful conclusions about happiness score from known tax : GDP ratio for a generic country.

I find:

raw happiness over tax:GDP ratio]

  • blue: OECD countries (to ease comparison with @Bobson's answer)
  • red: countries where happiness (+) or tax revenue : GDP ratio (x) is missing

With the unaided eye, I'd say that there's a maybe lower line of very roughly 1.75 happiness points + 9 ⋅ total tax revenue/GDP. Very few countries are below that:
apparent lower and upper limits

There may or may not be a similar upper limit as well at roughly 5.25 happiness points + 12 ⋅ total tax revenue/GDP (dark green line).
The five (dark green dots) countries above are the United Arab Emirates, Bahrain, Qatar, Kuwait and Saudi Arabia, i.e. the rich Arab oil countries that probably use lots of oil sales revenue instead of taxes.

And, anyways, even the most happy countries so far do not manage to get a happiness score of more than 7.6 (red line).

If I do a linear regression of happiness score over Tax revenue : GDP ratio for all countries where both happiness and tax revenue are available, I get the following picture:

linear regression

Regression formula is 4.425 happiness score + 4.36 ⋅ total tax revenue/GDP. Both intercept and slope are highly significant*, and the corresponding 95 % confidence interval for the regression line is the reddish shaded area in the graph. However, knowing that on average over almost 150 countries happiness correlates with tax revenue to GDP ratio (correlation coefficient is, btw, 0.49) doesn't help us saying anything about a single country. For that, we'd need to look at the prediction intervals (blue shaded area).

That tells us that for a generic country with 25 % of GDP going into taxes, we'd expect a happiness score somewhere between 3.5 and 7.4 (95 % prediction interval). Happiness score of 3.5 is what Guinea, Liberia and Togo in West Africa have (Syria is 3.46). 7.4 basically scatches the very top we actually have anywhere worldwide, it is between the Netherlands and Finland (topmost was Norway with 7.54). In other words, we'd expect it to avoid being one of the 6 most miserable countries, and also not to make it into the top 5 countries. Seeing that this is the 95 % prediction interval, so we'd expect about 5 % of such countries to be outside that range we just calculated. In summary, the variation between countries with similar tax to GDP ratio dominates.

A few more things to consider / thoughts:

  • Particular social insurance varies across countries not only in quality and costs, but also in how it is organized. And that latter can vary from 100% in government hand (and tax paid) over mandatory but separate from taxes and available if the individual wants to informal (family + friends) to basically unavailable for normal people. The first 2 possiblities can vary from 0 to 100 % tax money - even for the same percentage of GDP going into health care. For example, health care in Germany is to a large extent not by tax (i.e. government budget), but by compulsory health insurance fees. The extent of this is another 0.07 of GDP. Compare to Italy, where health care is financed by taxes. That is, the Italian tax : GDP ratio includes health care, whereas for Germany that comes on top.

  • So we do have a correlation. As we all know, that doesn't make a causation. And IMHO we can have plausible causation stories in both directions:

    • paying taxes makes people unhappy - they have less money which they can spend on what they need most.
    • That is, unless they perceive that they get back a sensible value of their taxes: if taxes are spent on things people in fact think important.
      This is something I've anecdotally heard e.g. Italians complain about: at 43.5 % of GDP they are taxed about as heavily as Germany - but they don't feel they get as much for their taxes (but see the health care difference above), in particular, they perceive that their government/administration is inefficient (we Germans do complain about that as well, but after spending some years in Italy I think we're in fact much better off)
    • low taxes can be a sign of various situations a country can be in, ranging from efficiency (Switzerland), i.e. being able to supply service without needing too much taxes to governments that are unable to collect taxes which may be a symptom of general inefficiency.
    • Maybe the lower line of happines vs. tax : GDP ratio indicates that if the heavy taxation doesn't lead to happiness, people are unwilling to pay those taxes...
  • Even if you look at high tax high happiness countries, you could have either:

    • taxes paid making people happy or
    • happiness making people pay more taxes (being OK with being taxed)

* regression for the OECD countries only does not yield a significant slope, and the same is also true for the non-OECD countries. But non-OECD countries have on average lower happiness scores and lower tax to GDP ratios.

| improve this answer | |
  • This is a great answer! I appreciate the research you put in (significantly more effort than I did to just google it a bit). It could probably use a TL;DR to sum it up, though. – Bobson Dec 27 '18 at 14:23

Analyzing comparative government policies is hard, since there's no way to have a control group or rigorous studies. All political scientists can do is analyze reported data and try to tease out correlations.

On this subject (as would be expected) there is no definitive answer, but it has been researched. For example, here is one article which discusses it, based on The Political Economy of Human Happiness by Benjamin Radcliff.

The article states:

Indeed, the evidence assessed by Radcliff suggests [a link between government spending and happiness]. His data shows that for one of the metrics, linked to welfare spending, countries scoring high on this indicator, happiness levels are above one point higher than low-scoring countries. He suggests that this contribution to happiness is double that of being married (being married is positively correlated with happiness), and three times the negative drag of unemployment.

However, it also includes this graph, which shows a very weak correlation:

Tax revenue vs happiness level

The graph here compares taxation levels (tax revenue as % of GDP) with happiness levels (life satisfaction), based on data from the OECD and the World Happiness Report quoted above. It shows an increasing trendline, associating a level of taxation of 20% in this group of OECD countries with a happiness level of around 6.5. All others thing equal, a level of 50% is correlated with a happiness level of around 6.8: some one thirds of a point higher across the trendline.

But not all others things are equal: the distribution is broad and the effects are very diverse. Denmark is on the top right with a happiness level of 7.526 and the very highest tax level of 50,9. On the far left, we find Switzerland with a marginally lower happiness level of 7.509 and only half the tax rate at 26.6%. On the lowest part of the graph, with happiness levels just above 5 points, we find Portugal, Greece and Hungary, with taxation levels around 34-38%.

So there is research on the subject, but no definitive relationship.

| improve this answer | |
  • 1
    I'm sorry, but I find the combination of the graph (which doesn't reach statistical significance at 5 % level much less than anything that could be used for prediction) with the quoted part of the article highly misleading. In contrast to the answer as it is right now, the article explicitly discusses the very broad variation encountered in the graph. For now, -1. – cbeleites unhappy with SX Dec 26 '18 at 15:11
  • 1
    @cbeleites - Yeah... I don't know what I was thinking when I combined that quote with that graph. Now that I'm more awake, I've revised it. – Bobson Dec 26 '18 at 16:25
  • What this graph tells me is: there is no correlation. – user253751 Dec 28 '18 at 4:57

Taxation for most people is giving what they need (their money) and in return largely receiving what they do not need. If anything the more government spends, the lower quality of life a people will have.

| improve this answer | |

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .