Schlingemann et al (2020) compared the share of employment and GDP attributed to stock market companies in the US in the 1970s and in 2019. Both measures of economic relevance have decreased over time, partially because of the switch from manufacturing to services. Companies in the service sector are less likely to be listed and, consequently, their increased economic significance has decreased the stock market’s representativeness.
Introduction
Last summer, the US had its lowest employment rate since 1948 while the NASDAQ and S&P 500 reached their highest values to date. The dramatic contrast emphasizes the disconnect between the stock market’s performance and general economic activity. The paper frames the dissonance within the historical context of a switch from manufacturing to services. As the economy transforms, the stock market’s degree of general market representativeness decreases as well. In the discussed paper, representativeness is defined as the fraction of GDP and employment provided by public companies.
Methodology
The paper estimates the listed companies’ contribution to US GDP and employment from the 1970s to 2019. It uses approximations because listed companies usually only disclose their employee numbers and employment contribution at the global level, and do not provide country splits. Consequently, the paper uses the companies’ global employment numbers and approximates their value-added according to previous relevant research. This methodology leads to an overestimation of the public companies’ contribution to the US economy, as not all their workers are located in the US. In fact, their contribution is more inflated in 2019 compared to the 1970s because companies have become more international over time. As a result, the paper’s estimate of the change in stock market representativeness in 2019 compared to the 1970s should be treated as an upper limit of the real value.
Results
Schlingemann et al (2020) found that listed companies contributed a lower share of GDP and employment in 2019 compared to the 1970s. More specifically, the fraction of non-farm private share workers that were employed in public companies were estimated to have dropped from 41% in 1973 to 29% in 2019.
The lower contribution is partially attributed to the US economy’s switch from manufacturing to services, where service companies are much less prone to be listed than manufacturing ones (see Figure 1). The difference reflects a dissimilarity in the incentives they face. For example, service companies are generally much less capital intensive than the industrial ones and are therefore less interested in the capital benefits the stock market can provide them.
Figure 1: 80% of manufacturing companies were publicly listed in the US in 2019
DATA: US Bureau of Labour Statistics (BLS), Bonsai Economics.
Does the stock market have to represent the economy?
Despite the apparent disentanglement between stock market performance and general economic activity in the US, it is important to highlight that from the beginning there was no compelling theoretical argument about why the stock market must represent total economic activity. Some other examples of why it does not need to be the case are the following:
Listed companies are likely to be larger and operate within different market segments than unlisted companies. Consequently, their performance may not be well correlated.
Market capitalization partially reflects expected future earnings which are not reflected in the company’s current employment and GDP contribution.
Market capitalization may increase without an increase in employment and GDP contribution if the company produces the same good with a lower labour cost.
For the above, and other reasons, the extent to which the stock market represents total economic activity is an empirical question, which the Schlingemann et al (2020) paper investigated.
Comments