This Appendix describes the basic properties of the forecasts of unemployment and real GDP and a robustness check of the results reported in the Section III of main text of the paper.
Ball, L., Daniel Leigh, and Prakash Loungani, 2013, “Okun’s Law: Fit at Fifty,” NBER Working Paper No. 18668, (Cambridge, Massachusetts: National Bureau of Economic Research).
Barnes, Michelle, Fabià Gumbau-Brisa, and Giovanni Olivei, 2013, “Do Real-Time Okun’s Law Errors Predict GDP Data Revisions?” Working Paper 13-3, Federal Reserve Bank of Boston.
Blinder, Alan, 1997, “Is There a Core of Practical Macroeconomics that We Should all Believe?” American Economic Review, Vol. 87 (May), pp. 240–43.
Fendel, Ralf, Elisa Lis, and Jan-Christoph Rülke, 2011, “Do Professional Forecasters Believe in the Phillips Curve? Evidence from the G7 Countries,” Journal of Forecasting, Vol. 30 (March), pp. 268–87.
Freeman, Donald, 2001, “Panel Tests of Okun’s Law for Ten Industrial Countries,” Economic Inquiry, Vol. 39 (October), pp. 511–23.
Lee, Jim, 2000, “The Robustness of Okun’s Law: Evidence from OECD Countries,” Journal of Macroeconomics, Vol. 22 (April), pp. 331–56.
Mitchell, Karlyn and Douglas Pearce, 2010, “Do Wall Street Economists Believe in Okun’s Law and the Taylor Rule?” Journal of Economics and Finance, Vol. 34 (April), pp. 196–17.
Moosa, Imad, 1997, “A Cross-country Comparison of Okun’s Coefficient,” Journal of Comparative Economics, Vol. 24 (June), pp. 335–56.
Okun, Arthur, 1962, “Potential GNP: Its Measurement and Significance,” Proceedings of the Business and Economics Statistics Section, American Statistical Association, pp. 89–104.
Paldam, Martin, 1987, “How Much Does One Percent Growth Change the Unemployment Rate? A Study of 17 OECD Countries, 1948-1985,” European Economic Review, Vol. 31, No. 1-2, pp. 306–13.
Pierdzioch, Christian, Jan-Christoph Rülke, and Georg Stadtmann, 2011, “Do Professional Economists’ Forecasts Reflect Okun’s Law? Some Evidence for the G7 Countries,” Applied Economics, Vol. 43, No. 11, pp. 1365–73.
Rülke, Jan-Christoph, 2012, “Do Professional Forecasters Apply the Phillips Curve and Okun’s Law? Evidence from Six Asian-Pacific Countries,” Japan and the World Economy, Vol. 24 (December), pp. 317–324.
Silvapulle, Paramsothy, Moosa, Imad and Mervyn Silvapulle, M, 2004, “Asymmetry in Okun’s Law,” Canadian Journal of Economics, Vol. 37, No. 2, pp. 353–374.
Sögner, Leopold and Alfred Stiassny, 2002, “An Analysis on the Structural Stability of Okun’s Law—A Cross-country Study,” Applied Economics, Vol. 34, No. 14, pp. 1775–87.
Johns Hopkins University and International Monetary Fund, respectively. The authors are grateful to Hites Ahir, Angela Espiritu and Ezgi Ozturk for excellent research assistance. We also thank Amy Guisinger for valuable help and comments on previous drafts of this paper.
Fendel, Lis and Rülke (2011) provide evidence, using data from the same source as the one used in this paper, that unemployment and inflation forecasts for G7 countries are consistent with a belief in the Phillips curve. Mitchell and Pearce (2010) find that interest rates “responded more in accord with the Taylor Rule than is evident in the predictions” of United States forecasters.
Tillmann (2010) finds that “individual forecasts for output growth and unemployment submitted by FOMC members suggest that the link between these two variables weakened significantly” in the 1990s, which he suggests arose because “policymakers were aware of a change in labor productivity” over this period.
It may be preferable to use the early releases of the unemployment and GDP data (since it is likely that these are closer to the objects that forecasters are trying to predict) rather than the final revised data. In practice, however, the choice of the early release (or ‘real time’) data vs. final data does not make a big difference, as Guisinger and Sinclair (2014) show in their comment on this paper. Similarly, Pierdzioch and others find that estimates of the Okun’s Law for the U.S. are similar whether one uses the final data or the real time data from the Federal Reserve Bank of Philadelphia database.
Both unemployment and real GDP forecasts show negative bias (that is, forecasts over-estimate both unemployment and real GDP growth). Moreover the bias in unemployment declines from h=24 to h=10 and then increases; while this pattern is difficult to explain, the magnitude of the bias is quite small. Explaining these biases is not the focus of the paper and they do not affect the reliability of the main results.
Ball and others (2013) report estimates of the “gap” version of Okun’s Law for 20 advanced economies, including the nine studied in this paper, for the period 1980 to 2011.
Rülke (2012) uses individual-level forecasts from Consensus Economics for six Asian-Pacific countries (three of which—Australia, Japan and New Zealand—are in our sample) and concludes that “professional forecasters believe in … Okun’s law. This result … is robust when using time-varying coefficients, different forecast horizons and when taking business cycle asymmetries into account. The results also suggest that the confidence in [Okun’s Law] is more pronounced during the economic crisis 2007–2009 and when looking at longer forecast horizons. Interestingly, the coefficients based on the actual series are similar to those based on the forecasts.”
An equation specification without a constant term does not alter our results. The results are also robust to the choice of forecast horizon, i.e., changing the months over which the final and initial revisions are calculated.
In preliminary work, we estimated a two variable VAR consisting of (1) monthly revisions to forecasts of unemployment changes and (2) the monthly revisions to real GDP growth forecasts. The impulse responses show that not only do unemployment forecasts respond to an innovation in real GDP forecasts but also that real GDP forecasts respond to innovations in unemployment forecasts. Thus there indeed seems to be a bi-directional relationship that needs to be explored further. Barnes, Gumbau-Brisa and Olivei (2013) find that, in the data, realtime errors in Okun’s Law contain information about future revisions to GDP. It would interesting to see if forecasters are aware of this relationship and use it to revise their GDP forecasts.