Cloudy with a chance of tariffs: the impact of policy uncertainty on the global economy

Ed Manuel, Alice Pugh, Anina Thiel, Tugrul Vehbi and Seb Vismara

Average tariffs on goods traded between the US and China increased by 15 percentage points from early 2018 to 2019. By making it more costly to buy goods from abroad, higher tariffs have reduced global trade flows and spending by households and businesses. But ‘direct’ effects of tariffs are not the only ways in which trade-related issues can affect global growth. Trade-related uncertainty has risen sharply since the escalation of trade tensions in 2018, which may have caused businesses to postpone costly investment decisions and financial conditions to tighten. In this post we investigate the size of these ‘indirect’ channels.

Higher uncertainty depresses business investment

Uncertainty has risen since the onset of trade tensions in 2018. Chart 1 shows two text-based measures of uncertainty produced by Baker, Bloom and Davis, which quantify newspaper coverage of economic and policy-related uncertainty. The orange line shows the evolution of uncertainty in the US, while the blue line shows developments in uncertainty globally for the period 1997–2019. Both measures have risen since tariffs started to increase in 2018. 

In the first stage of our analysis, we quantify the effects of higher uncertainty on US business investment. We estimate a vector autoregression (VAR) model that contains three variables: one of the uncertainty measures shown in Chart 1, and the quarterly growth rates of US business investment and GDP. Results from the VAR show that a one standard deviation rise in uncertainty reduces US business investment by 1% after a year (Chart 2). The results are robust to the choice of uncertainty measure including a trade-only measure of uncertainty.

The VAR specification allows us to calculate the contribution of the rise in uncertainty to US business investment growth since 2018. We find that elevated uncertainty has been an important driver of US business investment over that period. By the end of the second quarter of 2020, business investment is estimated to be 5%–7% lower than it would have been without the rise in uncertainty.

Chart 1: Uncertainty has risen since 2018

Note: US economic policy uncertainty measure is constructed using media citations of terms related to economic policy uncertainty. Global economic policy uncertainty measure is constructed similarly based on GDP-weighted average of the 21 national economic policy index values.

Source: ‘Measuring Economic Policy Uncertainty’ by Scott Baker, Nicholas Bloom and Steven J. Davis at

Chart 2: Response of US business investment to uncertainty

Note: Exogenous movements in uncertainty are identified using a recursive (Cholesky) ordering. We order the uncertainty variable first in the model, followed by the investment and GDP. The model uses quarterly data and is estimated for the period 1990 Q1–2019 Q2.

Effect of uncertainty on global growth

The results in the previous section tell us how much uncertainty has depressed investment spending in the US, however we want to assess the effects of uncertainty on global growth. This will depend not just on the direct effects of uncertainty on investment within a given country, but also on ‘spillovers’ between countries: weaker investment spending will reduce demand for both domestic and foreign goods, meaning that other countries will experience lower demand for their exports.

To calculate these global spillovers, we start with the finding from the previous section that the rise in uncertainty since 2018 will reduce US business investment by 7% in total. We assume that China experiences the same hit to investment from uncertainty, and that uncertainty also weighs on investment in the euro area, although to a lesser extent. Businesses in some euro-area countries rely heavily on external demand, and annual business investment growth has fallen from around 4% in mid-2018 to around 2% in 2019 Q2. 

We calculate the spillovers from this reduction in business investment using the National Institute’s Global Econometric Model (NiGEM), a trade-linked model covering 60 countries and regions. Based on the historical correlation of variables, this allows us to track how shocks to one country impact the rest of the world including via trade and financial channels. In aggregate, we find that the reduction in business investment has reduced the level of PPP-weighted world GDP by 0.4% by 2019 Q2. (In other words, the effects of uncertainty on investment have accounted for around 1/3 of the global slowdown.) That effect on global growth rises to -0.7% by 2022 Q2, assuming uncertainty remains elevated at its pre-Covid elevated levels observed in mid-2019 (Table A).

Table A: Peak uncertainty impacts on the level of GDP in per cent

By 2019 Q2Total (2019 Q2 to 2022 Q2)
United States-0.6-1.1
Euro area-0.3-0.5
Emerging markets, excluding China-0.2-0.4
World (PPP)-0.4-0.7
World (UK)-0.3-0.6

The risk premium effect

Besides the effects of trade-related uncertainty on investment, trade tensions can affect financial asset prices. Suppose market participants become less confident about companies’ ability to generate profits in future. In response, equity risk premia – the compensation investors require for holding equities rather than less-risky government bonds – and corporate bond spreads could rise. Market participants might also allocate a higher proportion of their investments to safer assets, such as government bonds. This could serve to push down on term premia, the compensation investors require for holding longer-term bonds rather than less-risky shorter term government bonds.

To assess the effects of trade tensions on asset prices between 2018 and 2019 Q4, we conducted an ‘event study’ of major trade announcements using the following steps:

  • First, we calculated the change in 10-year US Treasury yields in the 30 minutes following each trade announcement since the start of 2018. These include announcements containing relevant trade news but where tariffs did not actually change.
  • Second, we calculated the proportion of the overall move in 10-year US Treasury yields during that day accounted for by this 30-minute reaction. We selected only those days when trade news formed a significant part of the overall daily movement. Such days were identified based on two subjective criteria. In the first, trade news must explain 50% or more of the overall daily move with at least 70% of the trade-related move persisting by the end of the day. In the second, such thresholds were lowered to 30% and 50% respectively.
  • Third, we summed all of the trade-related movements in asset prices occurring on these ‘significant’ news days to calculate the overall effect of trade-related news on asset prices since the start of 2018. In the spirit of model averaging, we used the average of the results identified using the different criteria.

We find that trade news has raised both equity risk premia and corporate bond spreads in the US between early 2018 and end-2019 (Chart 3). Over this same period, US term premia have fallen in response to trade news, consistent with investors retreating from risky to safer assets in response to the escalation in trade tensions. Repeating the same exercise, we find that commodity prices have also fallen in response to trade news (Chart 4). This may partly reflect the direct impact of tariffs on demand for targeted goods – for example, US-China tariffs on agricultural goods have reduced demand, causing prices to fall. In addition, investors tend to retreat from commodities in response to negative global growth news and during periods of heightened risk aversion (see Bernanke (2016)).

Chart 3: Response of US 10-year term premia, US Investment-grade credit spreads and S&P 500 equity risk premia to trade news (2018–19 Q4)

Chart 4: Response of commodity prices to trade news (2018–19 Q4)

Finally, we use NiGEM to investigate how trade-related movements in financial asset and commodity prices have affected global growth. We find that the impact of commodity price falls has been small. And the rise in corporate bond spreads and equity risk premia have broadly offset falls in term premia. As a result, global GDP growth appears to have been little affected by trade-driven moves in financial markets over this period.


Prior to the Covid-19 pandemic, trade policies, including increased tariffs, have reduced global growth and played an important role in the global economic slowdown that occurred prior to the Covid-19 pandemic. We show that higher trade-related uncertainty during 2018–19 has been an important factor behind this slowing in growth, as businesses have held back costly investment decisions. By contrast, the effects of trade uncertainty on financial markets and commodity prices have been much smaller.

Ed Manuel, Alice Pugh, Anina Thiel, Tugrul Vehbi and Seb Vismara work in the Bank’s Global Analysis Division.

If you want to get in touch, please email us at or leave a comment below.

Comments will only appear once approved by a moderator, and are only published where a full name is supplied. Bank Underground is a blog for Bank of England staff to share views that challenge — or support — prevailing policy orthodoxies. The views expressed here are those of the authors, and are not necessarily those of the Bank of England, or its policy committees.