Real wage and labour share developments in Sweden.

Spoiler: Analysing inequality in Sweden by looking at developments of capital and labour shares 1960-2019 show that inequality is not a big issue. Labour’s share in GDP is closely related to the real wage. Growth of hourly real wages has not lagged labour productivity growth.

In the previous post, I showed that the wage share had increased in Sweden since 1993. That conclusion may be based on the initial point in the sample. In fact, it wasn’t. The wage share has been, as Kaldor suggested a long time ago, relatively constant over a long period of time. In this post, I will take a longer view on capital and wage shares in Sweden.

Even though the wage share appears to have declined slightly since 1960…

Let’s begin by having a look at developments since 1960. Labour share developments seem a bit volatile especially during the 1970’s. Periods of overheating of the economy and the labour market were followed by devaluations of the exchange rate during this period. The Swedish Crown was devalued five times between 1976 and 1982. Finally, the fixed exchange rate was abandoned in 1992 following our homemade financial crisis. After this turmoil, the wage rate has recovered and has increased since the 1990’s. Its 2019 level is merely two percentage points below the 1960 level. For the whole period, the net capital share decreased by five percentage points while the depreciation share increased seven percentage points, c.f. Figure 1.

Figure 1. Wages, net profits, and depreciation of the capital stock as shares in GDP 1960-2019 in Sweden.

Source: Ameco Database, DG Ecfin, European Commission.  Note: Capital and wage shares are adjusted for imputed compensation of self-employed.

...it has actually increased when only taking pure factor incomes into account.

The depreciation share starts to increase a few years before the year 2000. This is reflected in a falling net capital share. These developments imply that wages as a share of pure factor incomes, increase.  To find out, depreciation is subtracted from GDP at factor costs below. This adjustment shows that the Swedish capital income share has decreased from 24% in 1960 to 20% in 2019. The Swedish wage share increased from 76% to 80% during the same period, c.f. Figure 2.

Figure 2. Wage and capital incomes as shares in factor costs minus depreciation 1960-2019 in Sweden.

Source: Ameco Database, DG Ecfin, European Commission.  Note: Capital and wage shares are adjusted for imputed compensation of self-employed.

Falling shares of labour in GDP are often, but not always, reflected in real wage developments lagging labour productivity growth. The relationship can be illustrated by economic theory. As I am a lazy sod, I will simplify matters. In markets characterised by monopolistic competition, prices are set as a mark-up over marginal costs. After some simple manipulations, which can be found in many textbooks, the real wage is expressed as a function of two ratios. The first ratio relates the importance of labour to the mark-up and the second production over labour, i.e. labour productivity. This means that real wages will increase if 1) the importance of labour increases (the mark-up decreases) and/or 2) labour productivity increases. The expression can also be used to show that the labour share will increase if 1) occurs.

More elaborated discussions would also mention elasticities of substitution of capital and labour, capital labour ratios and relative factor prices. See for example here and here. As this study emphasises, the behaviour of the economy-wide labour share veils sectoral variations.  

The relative constant (or slightly increasing) labour share is reflected in close co-movements of real wages and labour productivity

Having the above-mentioned complications in mind, the relative constant labour share implies that with the exception of the volatile 1970’s, growth of real wages and labour productivity growth have moved together. raphing these variables and indexing them to equal 100 in the year 2000, shows that real wages growth exceeds labour productivity growth in since 2000. It remains to be seen how long this development can last, c.f. Figure 3.

Figure 3. Real hourly wages and labour productivity 1960-2019 in Sweden.

Source: Ameco Database, DG Ecfin, European Commission.  Note: Labour productivity is defined as GDP per hours worked.

My ramblings above confirm that income inequality in Sweden is not a big problem which I wrote about here.

Inequality can be discussed in other terms than income and wealth. Poverty and material deprivation gauge aspects of inequality that gauges the inabilities of households to pay unexpected expenses, afford a holiday away from home and more. That will be covered in a later post.

A word of a caution.

One should always question the choice of the base year for indices. A quick look at the figure indicates that if 1980 was chosen as the base year, real wages and GDP per hours worked would be much closer to each other in the end of the period. Choosing 1960 as a base year would, on the other hand, increase the gap between real wages and GDP per hours worked. Two things remain unchanged; hourly real wages exceed labour productivity in the 1970-1990 and during the last ten years or so, c.f. Figure 4.

Figure 4. Real hourly wages vs labour productivity 1960-2019 in Sweden.

Source: Ameco Database, DG Ecfin, European Commission.  Note: Labour productivity is defined as GDP per hours worked. The blue 45 degree line indicates where real wages equal labour productivity.

Anyway, both series move around the trend line implying that they form a stationary relationship but this furry guy doesn’t care.