Several hypotheses have been put forward in recent years to explain the downward trend of the labor share of GDP, with most of the explanations focusing in one way or the other on the bargaining power of labor.
It has been suggested, for example, that the erosion of labor unions, which started during the Reagan/Thatcher years, has put downward pressure on the wage share by decreasing labor's bargaining power.
A similar argument can be made for globalization and financialization. As the former communist economies joined the capitalist system, billions of workers were added to the global labor pool, thus putting downward pressure on unskilled labor in advanced economies, which now faces increasing competition from abroad.
Financialization usually refers to the fact that the financial sector has become increasingly important as a share of the domestic economy, which again seems to have put some downward pressure on the labor share of income.
More recently, some authors have suggested that the increase in the capital share is related to rising monopoly power and increasing markups. More specifically, some sectors are increasingly concentrated. This is especially true for the new economy where a few so-called superstar firms, such as Apple, Google, and Facebook, are dominating their respective industries. More importantly, those companies employ a large amount capital and relatively little labor. Think about the fact that the market capitalization of Facebook is about 500 billion dollars with more than 20.000 employees. Compare that to General Motors, which employs more than 200.000 employees but has a market capitalization of about 60 billion dollars only.
While most of the jobs in the high-tech sector are paying extremely well, the amount of jobs available are just a small fraction of all jobs available in the economy. However, the amount of capital employed and the profitability is much higher than in many other industries.
Last but not least, some economists have argued that increasing asset prices, especially real estate, are associated with an increase in the capital share. I have written about this topic myself on the website The conversation: https://theconversation.com/four-charts-which-should-worry-you-about-rising-house-prices-and-inequality-74707.
For a summary and literature overview of the academic papers you can have a look at my Master thesis:
https://lup.lub.lu.se/student-papers/search/publication/4617196
While inflation-adjusted house prices have stayed relatively stable for a long time period from the end of the 19th century until the middle of the 20th century, there was a veritable regime change that started a couple of decades after World War II. Many advanced economies liberalized their financial system and experienced significant run-ups in real house prices, especially over the last two decades or so. In most countries the house price booms were accompanied and probably exacerbated by significant increases in household debt, the result of mortgage expansions. High house prices should ultimately also support higher rents, thus leading to an increase in rental income. However, the homeownership ratio in many countries is quite high.
So now we have to get into the nitty-gritty of GDP statistics. The layman's definition of GDP is simply that it is the sum of all expenditures in an economy. Alternatively, one person's expenditure is another person's income, thus GDP is also the sum of all incomes. However, there are some technical adjustments that are made to calculate total GDP. More specifically, it should be noted that imputed rents as well depreciation are also part of GDP. the imputed rent is defined as the rent a homeowner would pay to himself in rent for his own property. Why exactly is this included in GDP? The answer is to avoid distortions. Imagine two economies that are similar in every other aspect. And let's assume for simplicity that there are only two persons, Frank and Jack. In economy A, both Frank and Jack bought a house in which they live and they thus do not have to pay any rent. In economy B, it just so happens that Frank lives in Jack's house and thus has to pay rent to Jack and also vice- versa, so Jack lives in Frank's house and thus has to pay rent to Frank. GDP in the economy B will be higher (inflated) simply as a result of the two person's paying rent to each other and not because economy B is more productive. In order to avoid inflated GDP numbers for economies with low homeownership ratios, such as Germany, the US statistics office calculates imputed rents for homeowners and includes them in the GDP figures. The imputed rent of a house depends on its size, quality, location, etc. In short, imputed rents are ultimately a function of the market price of the property in question. Large run-ups in house prices should this ultimately translate into higher imputed rents. Depreciation of capital (machinery) is also included in the GDP figures, the reason being that depreciation is counted as cost of production.
The figure below depicts total employee compensation as a share of GDP. The numbers are from the BEA. While one can clearly see the recent downward trend in labor income, it should be noted that the labor share is not significantly lower than in the aftermath of World War II. Instead, it rather seems that the 1970s were a little bit of an outlier in the sense that the wage share was extraordinarily high.
Wage compensation / GDP
Depreciation / GDP
By the way, this also implies that a large part of the academic literature might be wrong about the causes of a falling wage share.