Lewis-Mertens-Stock (NY Fed) weekly economic indicators and Baumeister, Leiva-Leon and Sims WECI and Woloszko (OECD) weekly trackers Weekly indicators to 11 February 2023 were published today.
figure 1: Lewis-Mertens-Stock weekly economic index (blue), OECD weekly tracker (tan), Baumeister-Leiva-Leon-Sims U.S. weekly economic conditions index plus 2% trend (green) and IHS Markit monthly GDP (blue bars).Source: New York Fed via fred, OECD, WECIand the authors’ calculations.
The Weekly Tracker rebounded sharply, dipping into negative territory for the week ended 26th November, and now surpasses WEI (0.9%) and WECI+2% (2.0%). A WEI reading of 0.9% for the week ended 11 February 2011 could be interpreted as a 0.9% YoY increase if the 0.9% reading persisted throughout the quarter. Baumeister et al. A reading of -0.03% is interpreted as a growth rate of -0.03% above the long-term trend growth rate. US GDP growth averaged around 2% over the period 2000-19, so this translates to an annual growth rate of 1.97% as of 11 February 2011. The OECD weekly tracking index reading of 3.4% can be interpreted as a year-on-year growth rate of 3.4% as of 11 February 2011.
The OECD Weekly Tracker continues to rise even as the other two series slowly decline. It is important to remember that WEI relies on correlations of ten series available at weekly frequencies (e.g., unemployment claims, fuel sales, retail sales), while WECI relies on mixed frequency dynamic factor models. The Weekly Tracker — at 1.1 percent — is a “big data” approach that uses Google Trends and machine learning to track GDP. As such, it does not itself depend on actual economic indicators.
The “Weekly Tracker” shows a more direct decline in the 2020 recession than measures based on real economic indices.
figure 2: Lewis-Mertens-Stock weekly economic index (blue), OECD weekly tracker (tan), Baumeister-Leiva-Leon-Sims US weekly economic conditions index plus 2% trend (green). Recession dates as defined by NBER are shaded gray.Source: New York Fed via fred, OECD, WECINBER, and author’s calculations.
The dynamic factor approach is more familiar to me, so I think at the moment I trust WEI and WECI more (that is, the Weekly Tracker is more in line with the IHS Markit/S&P Global monthly GDP y/y growth as shown this post).
