The Unassuming Economist

A new paper finds that “Gross domestic income and gross domestic product—GDI and GDP—measure aggregate economic activity using income and expenditure data, respectively. Discrepancies between the initial estimates of quarterly growth rates for these two measures appear to have some predictive power for subsequent GDP revisions. However, this power has weakened considerably since 2011. Similarly, the first revision to GDP growth has less predictive power in forecasting subsequent revisions since 2011. One possible explanation is that evolving data collection and estimation methods have helped improve initial GDP and GDI estimates.”

David Mihalyi and Tommy Morrison at NRGI created a World Economic Outlook Forecast Tracker that enables users to see how IMF economic projections have evolved over time. On it, you can select from an expansive list of countries and country groupings to track how IMF forecasts evolved year-to-year for dozens of economic indicators, such as GDP growth, government revenues and the budget deficit as well as the price of various commodities. The app shows an animated plot of the forecasts and historical values over 10 years, as well as providing a data download and a plot download (example attached).

From a new post by Julia A. Seymour :

“New York Times economist Paul Krugman immediately reacted to the 2016 election of Donald Trump by warning of a possible “global recession.” Perhaps Yahoo! was taking pointers for its latest series. Even though the economy has been doing well, Yahoo! Finance just launched “Your Next-Recession Survival Guide” warning it is “time to prepare for the economic downturn, which could occur as early as 2020.” The new series began June 20.

[…]

In general, forecasting is unreliable. Financial Times wrote in 2014 that an analysis of all 1990s economic forecasts concluded there was great similarity between them and “the predictive record of economists was terrible.” Prakash Loungani, the author of the study, said “The record of failure to predict recessions is virtually unblemished.” ”

My paper is available here.

From a new article from the The Economist:

“The Economist has built a statistical model to identify what makes a country good at football. Our aim is not to predict the winner in Russia, which can be done best by looking at a team’s recent results or the calibre of its squad. Instead we want to discover the underlying sporting and economic factors that determine a country’s footballing potential—and to work out why some countries exceed expectations or improve rapidly. We take the results of all international games since 1990 and see which variables are correlated with the goal difference between teams.” “Our model explains 40% of the variance in average goal difference for these teams.”

From a new Forbes article by Raul Elizalde:

“Forecasting the economy is just as difficult as forecasting the stock market. Economists are very good at explaining what already happened and why, but not so at predicting what will happen next.

They know this. Prakash Loungani, an economist at the IMF, showed in a study that professional forecasters missed 148 out of 153 world recessions. This is not surprising: Economic indicators very rarely flash any warnings before a recession actually arrives. Economic downturns seem to come unexpectedly.”

My paper is available here.