Syndicate content


Quantifying uncertainties in global growth forecasts

Franziska Ohnsorge's picture
Figure 1. Risks to Global Growth
Upside risks to global growth have increased since January while downside risks for current-year growth have reached post-crisis highs.

A 90% confidence interval implies a 90% chance of growth falling within the given range
Source: World Bank Global Economic Prospects report June 2016.
Note: “90 percent JAN16” is the 90 percent confidence interval of a fanchart based on data available for the January 2016 Global Economic Prospects report

Assessing economic forecast uncertainty and the balance of risks to the growth outlook is critical to effective policymaking. Lower-probability but high-impact events can lead to significant deviations from baseline projections, and this  should be factored into policy design. The World Bank’s most recent Global Economic Prospects unveiled a tool to quantify uncertainty around global growth forecasts and presented it in the form of a fan charts (Figure 1)

The approach adopted in the Global Economic Prospects report consists of two steps.

First, a number of measurable risk indicators that are typical sources of forecast errors for global growth forecasts are selected. Three were chosen: equity price futures, oil price futures and bond term spreads (the difference between short and long term interest rates). For instance, greater volatility in oil price futures could be associated with rising uncertainty around global growth forecasts, while a downward trend in equity price futures could signal rising downside risks to growth.

Second, the probability distributions of forecasts for these three indicators are then mapped to the distribution of global growth forecasts. Both the degree of uncertainty and the balance of risks to the forecast are approximated by weighted averages of the standard deviation and skewness implied by the distributions of expectations for the risk indicators. The weights are estimated in a vector autoregression model (Ohnsorge, Some, and Stocker 2016). To account for potential asymmetry in the distributions of risks, a two-piece normal distribution is assumed, in line with other studies.

Back to the Future

Eliana Cardoso's picture

Imagine if, in 1799 – the year in which Napoleon seized power – a research institute had published its global forecasts for the next 20 years. Its researchers would have known about the tremendous changes that took place over the previous two decades: from the United States’ declaration of independence, through the French Revolution and the execution of Louis XVI, up to Napoleon’s victory over Austria in his Italy Campaign.

Even so, the chances of the researchers accurately predicting the events that came to pass over the subsequent 20 years, including their impact on the 19th century’s world order, would have been infinitesimal. No one could have anticipated that Napoleon would have plunged Europe into non-stop war for a decade until being overcome at Waterloo, or that, by the time of his defeat, he would already have swept away the foundations of traditional structures and initiated an unstoppable wave of reforms.

Because of its industrial might, this Europe would dominate the rest of the world during the 19th century. When European rivalries exploded into World War One, the face of the earth had already changed considerably compared to the previous century. And, having changed the world, Europe set the conditions for the demise of its own empire. Even before World War One, Teddy Roosevelt had heralded the start of the United States’ ascension to its current hegemony.

China grew faster than its target and most projections in 2009 – what are the key takeaways?

Louis Kuijs's picture
Click image to enlarge.

China’s economy grew 8.7 percent in 2009. This was more than the 8 percent target, despite the global recession that caused global output excluding China to fall about 3 percent. China’s growth outcome is substantially higher than projections made in early 2009. For instance, in our  World Bank quarterly economic update (of which I am the lead author) we projected 6.5 percent GDP growth and some other forecasts were even lower (see Figure 1).

How did these forecasts come about, and what lessons we can draw from the experience of China’s growth in 2009? I cannot speak for my colleagues at the World Bank, let alone for other economists. But, all in all, while I have learned important lessons, I am not sure how differently I would see and do things if again presented with a situation like we were in a year ago.