I like to be optimistic! But, after 2009 crisis, I wonder if the IMF carry a happy bias in world GDP growth projections?
On Thursday, April 18, IMF launched chapter 1 of World Economic Outlook (WEO), one of its key anchor publications, with world GDP growth projections, where IMF forecast global growth of 3.5 percent for this year, and 3.6% for 2018. As we look at this and past IMF WEO publications, taking WEOs after the 2009 economic crisis, there appears to exist a happy bias, where world GDP growth projections carry an upward bias.
The theory of rational expectations was originally proposed by John F. Muth (1961) , which later became an anchor idea in economic modeling when it was used by neoclassical economists such as Robert Lucas, Jr. , Edward C. Prescott, Thomas J. Sargent, Michael R. Darby, Finn E. Kydland, Lawrence H. Summers, and Neil Wallace.  This is a hypothesis of economic science which states that predictions about the future value of economically relevant variables made by rational agents should not be systematically erroneous and that the errors are random, uncorrelated and with zero mean (white noise).
This idea on how to model expectations, known as the rational expectation hypothesis, accepts that the expectations of economic agents may be individually erroneous, but correct on average. Thus, although the future is not entirely predictable, it is assumed that agents' expectations are not systematically skewed and that they use all relevant information to form their expectations about economic variables, including information about past mistakes.
IMF world GDP growth projections serve many important purposes for policy makers in member countries, for the private sector and the public in general. To the extent that these projections may affect policies, economic and investment decisions, they may affect the eventual economic outcome. Predicting the economy is complex, it is an art more than a science, and the only certainty is uncertainty. And, in this respect, models should be frequently revised and adjusted so projections only carry a white noise and not a bias.
We know that predicting the economy’s next few years is one of the areas in which economists fail considerable. But, the value of better predictions is enormous. For example, if decision makers have better forecasts that a recession is on the corner, they might prepare in a better way to weather the storm, and eventually take measures in advance to counteract or minimize the recession, and ease others and save millions from adversity and unemployment. The aim of having white noises in economic forecasting should be an objective in top notch economic modeling.
The good news is that if we believe in rational expectations, the market already discounts any happy bias, and make its own assessments of the accuracy of the different world GDP projections.
 John F. Muth. (1961). "Rational Expectations and the Theory of Price Movements", Econometrica 29, pp. 315–335.
 Lucas, Robert (1972). "Expectations and the Neutrality of Money". Journal of Economic Theory. 4 (2): 103–24.
 Edited by Preston J. Miller (1994). The Rational Expectations Revolution: Readings from the Front Line. The MIT Press. ISBN: 9780262132978 and ISBN: 9780262631556.