Bank of America: BofA global growth model: Third-quarter resilience stronger than economists expected, but regional signals diverge

Whose expectations did global growth exceed?

BofA’s GLOBALcycle model indicates that global GDP grew at an annualized quarter-on-quarter rate of about 3.7% in the third quarter of 2026, above the 3.4% aggregate forecast by the bank’s economists for the countries covered by the model. BofA believes that, despite shocks including high US interest rates and the Iran war, hard data generally continue to support global growth.
The gap comes mainly from emerging markets: the model estimates annualized quarter-on-quarter growth of 4.6% in the third quarter, above the 4.3% forecast by BofA economists; the respective figures for advanced economies are 2.2% and 2.1%, which are relatively close.

Among advanced economies, the model’s assessment of the US is weaker than BofA economists’ forecast, while it is more optimistic about the euro area and Japan. However, BofA notes that the model has recently overestimated growth in the euro area and Japan, while US fundamentals remain strong, so the magnitude of these gaps should be downplayed.
The model’s signals are slightly stronger for most emerging-market countries, including China; but the differences are larger for Turkey and Mexico, with economists forecasting stronger growth in the former and the model forecasting stronger growth in the latter.
The model’s recent estimates of global growth have also been slightly above actual outcomes. It persistently overestimated China’s growth after the pandemic; BofA attributes this to persistent domestic weakness and the suspension of employment data since 2022, and has therefore added a post-pandemic adjustment variable to its growth regressions for China and emerging markets.

The model combines data at different frequencies for each economy into an economic conditions indicator, then estimates GDP growth for the current quarter. Inputs include monthly retail sales, industrial production, employment or exports, as well as quarterly GDP; weekly equity indices are also used for most economies.
The coverage accounts for approximately 80% of global GDP on a purchasing-power-parity basis. The indicators for individual economies are aggregated using purchasing-power-parity GDP weights, which are updated on a rolling basis as the structure of the global economy changes.
BofA says historical tests show that the model tracks overall growth globally and across advanced and emerging economies more accurately than it does for most individual countries. However, the historical indicators were constructed retrospectively using the current methodology and incorporate the effects of hindsight design or adjustments.

BofA’s historical regressions show that stronger economic conditions indicator levels are typically associated with higher global equity returns, higher short-term US interest rates and a flatter yield curve, stronger emerging-market currencies and a weaker US dollar.
When growth momentum accelerates, the historical relationship differs: long-term US interest rates tend to rise and the yield curve tends to steepen. The positive correlation between momentum and oil and commodity prices is not statistically significant in the regressions.
These relationships explain only a limited share of the variation in asset returns, and their explanatory power has declined over time; BofA also finds that the regression coefficients are relatively stable.