Today, we are fortunate to present a guest contribution written by Paweł Skrzypczyński, economist at the National Bank of Poland. The views expressed herein are those of the author and should not be attributed to the National Bank of Poland.
In January 2019 I presented on Econbrowser a couple of probit model examples that enhance the standard U.S. Treasury yield curve slope model with the U.S. Economic Policy Uncertainty Index (EPUI) as measured by Baker et al. (2016). Today, after six months, I present an update that includes the data ending in June 2019. Figures below present the obtained U.S. recession probabilities in-sample and their out-of-sample forecasts. The predicted probabilities along with the forecasts made in January 2019 are reported in the figures and table below.
Figure 1: Recession probabilities from 10yr-3mo Treasury spread lagged 12 months (blue), and spread lagged 12 months augmented by 6 month lagged Economic Policy Uncertainty index (red). Source: author’s calculations.
Figure 2: Recession probabilities from 10yr-3mo Treasury spread lagged 6 months (blue), and spread lagged 6 months augmented by 6 month lagged Economic Policy Uncertainty index (red). Source: author’s calculations.
Predicted U.S. recession probabilities
|Specification||January 2019 forecasts||July 2019 forecasts|
|spread_12||23.4% in Dec-19||37.8% in Jun-20|
|spread_12_epui_6||39.0% in Jun-19||49.3% in Dec-19|
|spread_6||21.5% in Jun-19||30.3% in Dec-19|
|spread_6_epui_6||51.1% in Jun-19||56.0% in Dec-19|
Notes: Specifications of probit models for the NBER dated recessions binary variable on the lefthand side of the estimation equation are the “spread_12” – model with 12-month lagged 10Y3M spread as the only regressor, the “spread_12_epui_6” – model with 12-month lagged 10Y3M spread and 6-month lagged EPUI as regressors, the “spread_6” – model with 6-month lagged 10Y3M spread as the only regressor and the “spread_6_epui_6” – model with 6-month lagged 10Y3M spread and 6-month lagged EPUI as regressors.
Overall, things have gotten worse since January 2019 as the yield curve inverted and the EPUI remained elevated, which both contributed to the rise of recession odds over the next 6 to 12 months. Current estimates are depicted in Figure 1:
I believe that among others the ongoing uncertainty related to trade and Fed independence are here the key factors keeping the EPUI high, which in case of the presented models implies elevated recession odds.
Scott R. Baker, Nicholas Bloom, and Steven J. Davis. 2016. “Measuring Economic Policy Uncertainty.” Quarterly Journal of Economics, vol 131(4), pages 1593-1636.
Update, 7/5/2019 10:15AM Pacific:
I did a little bit of quick calculations assuming the EPUI over 2017:01 – 2019:06 equals the average EPUI during last 2 years of Obama (spreadsheet enclosed). Specification is spread_12_epui_6, parameters are the same as in case of the actual data set used for this model. Predicted counterfactual probability for Dec-19 is 18 percentage points lower.
Figure 3: Recession probabilities from 10yr-3mo Treasury spread lagged 12 months and 6 month lagged EPUI (red), and counterfactual holding EPUI over Trump administration equal to average over preceding two years (blue).f Source: author’s calculations.
This post written by Paweł Skrzypczyński.