Friday, April 26, 2019

Economists often get their forecasts wrong. Here’s why.

By Faris Mokhtar

Image result for monetary authority of singapore
26 April, 2019

SINGAPORE — Economists are paid to predict how the economy will perform — whether it will grow or shrink, and by how much.

But these professional forecasters do not always get it right. In fact, they can miss the mark by a wide margin.

In its latest macroeconomic review, published on Friday (April 26), the Monetary Authority of Singapore (MAS) waded into the issue of the difficulties that economists face when trying to predict a country’s gross domestic product (GDP) growth as well as upcoming economic recessions and recoveries.

Citing studies and media reports, it pointed out that forecasters, though aware that recession years are anomalies, tend to “miss the magnitude of a recession by a wide margin until the year had drawn to a close”.

They also made the largest errors ahead of GDP contractions, according to an assessment by The Economist.

Having conducted its own study on the challenges of making forecasts, Singapore’s central bank came to this conclusion: Professional forecasters here, similar to those in other countries, “are not able to capture the onset of recessions or recoveries in a timely way, and the extent of output decline or expansion is also often misestimated, especially for year-ahead projections”.

The MAS cited three main reasons for this:


This could be due to several reasons, including the fact that data on the economy may only become available after long lags, or be of poor quality.

Another reason: Macroeconomic models are designed to work well “on average” but not during unusual periods such as recessions, which tend to occur because of events that are hard to predict.


Apart from minimising forecast errors, forecasters might have other objectives. For instance, they may be interested in presenting forecasts which allow them to maximise publicity, revenue or prestige, according to a study.

Another research finding was that forecasters “might also lack incentives to break away from the consensus”. Why? Because the reputational loss from being wrong may be high.

As such, they are prone to “herding” or a tendency to make similar forecasts as other fellow economists, instead of issuing an “outlier” forecast.


A study published in 1987 by American economist William Nordhaus, known for his work in economic modelling, argued that forecasters only revise their estimates “slowly and insufficiently” in response to new information.

Professor Nordhaus noted that “people tend to smooth their forecasts too much”.

“That is, we break the good or bad news to ourselves slowly, taking too long to allow surprises to be incorporated into our forecasts,” he added.

No comments: