Time-varying Granger causality between the stock market and unemployment in the United States

ABSTRACT

In this paper, we look at the connection between the stock market and the unemployment rate in the United States. Using a recent time-varying Granger causality framework covering the period from January 1960 to October 2020, tests reveal that lagged realizations of the stock market have predictive power regarding unemployment, and vice et versa, but that the predictive ability only occurs sporadically over time, particularly during ‘crash’ periods. These results are in line with the literature on the information spillover between finance markets and the real-life economy, with changes of causality across time.

KEYWORDS:

• Stock market
• unemployment
• time-varying causality
• United States
• JEL CLASSIFICATIONC32
• E44
• G14
• J60

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

¹ Some studies examine whether and how the search intensity data obtained from Google Trends contributes to nowcasting of the U.S. unemployment rate compared to the conventional AR model (like Nagao et al. (2019)).

² Pan (2018) proposes a brief review of the literature.

³ In addition, in order to strengthen the robustness of the results obtained, we are carrying out Granger causality tests, with a lag of order 3 with regard to the AIC criteria: F-Stat = 7,47*** for p→u and F-Stat = 2.24* for u→p.

⁴ It should be noted that we obtain similar results with a VAR model: $y_t=\left[u_t,p_t\right]$.

Additional information

Funding

This work was supported by the ARIANE CPER and MSH Lorraine.