Lavorgna (SMBC Nikko) and Millar (Barclays) said that economic activity can be better estimated through GDI. Both believe GDI may be a better indicator of an impending recession (USA Today):
figure 1: GDP (black bold), GDO (light blue bold), GDP+ (red), GDI (blue), all in 2017 billion USD seasonally adjusted exchange rates. The GDP+ level uses the GDP+ growth rate of the GDP iteration in the fourth quarter of 2019. Source: BEA 2023 Season 3 Edition 2, Philadelphia Fedand the author's calculations.
GDI is indeed much lower than GDP (Zero Hedge Assert that GDP will be revised downward to GDI). GDI did fall by 0.2% year-on-year; on the other hand, the month-on-month annualized growth rate was +1.5%.Anyway, what I read Economic Commission for Europe (2015) and Nalervik and Braun (BPEA, 2011), is that GDO (the arithmetic mean of GDP and GDI) is a more accurate measure of final GDP, with GDP regressing more to GDO than otherwise. GDO growth was indeed more modest than GDP growth (3.3% vs. 5.2%, seasonally adjusted quarterly).
GDP+ published by the Philadelphia Fed incorporates information on revisions to obtain a more accurate estimate of final GDP. One problem with comparing this series is that only growth rates are provided. I iterated the GDP+ growth rate at the GDP level in the fourth quarter of 2019 (when GDP and GDO match). The implied level of GDP+ is close to the level of GDO in the third quarter of 2023, but growth is still relatively moderate.
as Jacobs et al. (JBES, 2022) point out (see Jim's post on GDP++), the statistical method used in GDP+ makes the volatility of the estimated value of the underlying factors of GDP smaller than the volatility of the observed value. GDP++ does not include this requirement. From the third quarter of 2021 to the second quarter of the second quarter, GDP++ growth is in line with GDO.
From my perspective, I value GDO more than strictly GDI (or GDP). This means that economic activity is 1.25% lower than the expenditure-side indicator (GDP) indicates.
