Thursday 29 November 2018

Forecasters - What Do They Know?


So the government has produced its Brexit forecast and the Brexiteers are shouting from the rooftops that forecasts are wildly inaccurate, without checking on the veracity of their statement, nor the direction of and historical inaccuracy.

  1. The Office of Budget Responsibility forecasts are not 'wildly' inaccurate. Excluding the forecasts for 2008 and 2009, which were compiled by the Treasury (the OBR only came into existence in 3010) and affected by an unexpected event - the global financial crisis, which could not be forecast as it was caused by fraud - the difference between the OBR growth forecast and actuality has never been more than 1%.
  2. The actual outcome has consistently been worse than the forecast, not better.
  3. If the OBR is forecasting a 4% drop in GDP for May's Brexit and 9% for a hard Brexit, on the basis of probabilities and past accuracy, it will be 5% and 10% - possibly more since the forecasts are for a longer period than just one year and the error will be compounded over time.
  4. A forecast of 2% growth versus a 1% reality is a large percentage error, but not a large absolute error and is a consequence of using small numbers. Brexiteers can be guaranteed to quote percentage errors in the coming days.
  5. Brexiteers have yet to produce a forecast, and they won't; they will persist in their own Project Mindless Optimism with nothing to back it up, hoping to garner the reactionary forces of huffing and puffing outrage from those who don't bother to check anything and believe what they're told by demagogues who sound authoritative.



No comments: