One of those seemingly self-evident truths that seem to skip effortlessly off the tongues of the left-wing advocates of equality, is that equality of income is essential to societal stability and contentment. But is that really true?
The Better Life Index of the OECD aims to measure the relative life satisfaction (happiness) of citizens of mainly OECD countries. The point allocated to each country, calculated on the basis of a series of answers in a survey, reflects the subjective views of respondents. Dealing as the survey does with a purely subjective value, namely happiness, this is perfectly in order. The one thing that a person cannot second-guess, is another’s say-so about his or her own happiness.
Here is the raw data (from the most equal society to the least), including data on GDP per capita, a good proxy for average income levels:
|Country||Equality (Gini coefficient) after transfers & taxes||Life satisfaction||GDP per capita|
First, a few observations:
- The most unequal of the developed countries in the list, is the US. Yet it still has an LSI of 7.2, which compares well with that of the happiest society, Denmark, at 7.5.
- The US fares even better compared to highly equal Sweden (7.2) and Norway (7.4)
- It is a far happier society than a whole host of other more equal societies on the list, including Czech Republic (6.5), Slovakia (6.1), Austria (6.9) and Germany (7)
- Switzerland, which tops the pile in terms of happiness together with Denmark at 7.5, is dead average in terms of equality, with a Gini coefficient of 33.7, placing it 18th out of the 36 on the list.
Assuming that equality matters for purposes of happiness, the rather obvious explanation for these seeming anomalies is that the per capita incomes of the happier societies – such as average-equality US and Switzerland – are far superior to that of any of the other more equal, but unhappier, societies.
If we plot the LSI (happiness) of each country on a graph against the income equality of each, it is immediately obvious that there is no correlation between the two. In other words, mere equality does not make people happier:
|31||Hu||Sl,||Ko||Ger, Lux||Aus Ne|
|33||Fr||Be, Ire||Swi, Ca|
On the other hand, if we were to plot the LSI of each country against its per capita GDP, it is immediately apparent that there is a compelling correlation between the two. It confirms the somewhat un-PC point that money, up to a point, does indeed seem to buy happiness:
|45||Swe,Au. Ger, Be
|35||Ko, It||Sp X||NZ, Is|
That of course does not mean that it may not be better to be both more equal and wealthier.
The point to make however, is that of the two factors (per capita income and equality), per capita income is by far the most influential. This we can tell by breaking the countries into peer groups according to income:
PCGDP over R50000:
40 000 – 50 000:
30 000 – 40 000:
20 000 – 30 000:
Less than 20000:
In none of the income categories is there any correlation between equality and happiness.
The real question is, however, whether there is a trade-off between the two factors of per capita income and equality of income. In other words, does a country have to give up income in order to achieve equality, and vice versa?
There are two types of policy measures that improve equality. The first is applied before government takes tax money and spends it on grants and welfare in order to equalise incomes. That category includes measures such as policies to bring about greater equality in the labour market.
First, let us examine labour law (the quantified value of which appears in the first column in brackets) as potential equaliser of income. A comparison of the Gini coefficient of different countries before taxes and transfers shows that labour laws have no significant impact on inequality of income. So for example the US, with significantly less labour regulation than most European countries, compares well with several highly regulated countries that are normally held up as paragons of equality (such as Austria and Germany, for example). The list comprises a wide variety of labour law systems with hugely differing degrees of labour regulation among both equal and unequal societies, with little discernible variation or trend.
(labour market freedom)
|Unemployment||Immigrant % of population||Ed||Gini before T&T||Gini after T&T|
|South Korea||12 (4.51)||2.7||2.9||.865||.344||31.3|
|New Zealand||15 (8.66)||5.6||25.1||.917||.455||36.2|
|Hong Kong||5 (9.30)||3.3||38.9||.767||.537||47.5|
A related factor is the percentage of a country’s workforce that is covered by collective bargaining agreements. The bargaining coverage of countries is often cited as a factor that significantly brings about greater equality, but the table clearly disproves this notion. If labour law contributes to equality in these listed societies, clearly it must have to do with some other aspect of the law, or other factors.
The only other labour law factor that realistically be significant, would be state-imposed minimum wages. However, none of the 6 most equal societies above has such minimum wages – except for South Korea, which is the exception proving the rule.
As for non-law factors, the countries above that have low inequality have low unemployment too, provided they also have low levels of immigration (Korea, Iceland, Switzerland). Singapore, Hong Kong and the US (with much higher inequality), by contrast, have low unemployment but high immigration.
Low unemployment is self-evidently conducive to equality (poor people earning wages always have higher incomes than poor people who are unemployed).
Immigration has an impact as it often – at least temporarily – lowers the educational and skills level and language- and cultural cohesion of the population. (Germany in a way of course has the greatest “immigration” burden of all in that West Germany had to absorb an entire country barely two decades ago. It is fair to infer that that partly explains its relative inequality of income.)
The most equal society in the world (as measured by its pre-T&T Gini coefficient), namely Korea, confirms this thesis as it has both very low unemployment and very low immigration levels.
High levels of education improves productivity and thus salaries. This is well illustrated by New Zealand, the US and Australia, where indifferent unemployment rates and quite high immigration levels are somewhat corrected by very high education levels.
These and similar non-law factors, one intuitively feels, account almost entirely for the difference in pre-T&T Gini levels, leaving very little room for labour law as a differentiating factor.
Minimum wages and similar measures designed to improve conditions of employment, thus have virtually no impact on equality of income, whilst education and job growth most certainly do.
It is of course so that taxes and transfers may be used to improve equality of income.
To deal with this, I compare the proportion of the economy of a group of wealthy countries (excluding purely oil-based economies) that consists of state expenditure, that being a good proxy of taxes and transfers.
Government expenditure as a proportion of GDP: Wealthy countries, excluding oil-based economies
|Country||State expenditure% (economic freedom rank)||Per capita GDP||Gini after T&T|
|Sweden||58.1 (42)||45 143||25|
|Denmark||58.1 (22)||44 862||28.1|
|Iceland||58.1 (85)||43 393||28|
|Belgium||56 (52)||42 725||33|
|Netherlands||54.7 (30)||47 130||30.9|
|Austria||54.3 (31)||46 164||26|
|Germany||48.8 (29)||45 615||30.6|
|Canada||48.2 (9)||44 088||33.7|
|Australia||43.6 (12)||43 901||30.5|
|Switzerland||37.8 (4)||56 939||33.7|
|Luxembourg||37.5 (27)||91 048||30.7|
|US||19.9 (16)||54 629||41.1|
|Hong Kong||17 (1)||55 084||53.7|
|Singapore||16.3 (2)||82 763||46.4|
The trade-off between taxes and transfers on one hand and equality on the other is clear. The higher taxes and transfers are, the lower the inequality of income tends to be.
Another trade-off is equally clear however. And that is between government taxes and transfers, and per capita income.
Countries with a government expenditure percentage over 40 percent of GDP (down to Australia in the list), all cluster around a per capita GDP of just over $40 000. The rest, whose government spending is under 40% of GDP, are significantly better performers with an average per capita GDP of $68 092. The incomes of these countries increase exponentially in line with declining state spending.
The fair inference from all this is that taxes and transfers tend to undermine per capita income, which as we have seen, tends to correlate with happiness. Amongst already developed countries this may be less important. Amongst developing countries like South Africa it matters hugely.
South Africa is thus far better off aggressively pursuing employment and growth than equality, in the pursuit of the happiness of our people.
This has important implications. Policies that amount to tax and spend such as welfare grants, do not support income growth and happiness in the long run. The same applies to policies that indirectly have the same effect, such as employment equity laws, BEE and minimum wages.