This week’s forecasts – Mixing targets or wishful thinking


19 February 2018 10:55
People in a meeting

Early February was not short of forecasting controversy. In this edition we focus on Government estimates of the effects of different types of ‘Brexit’ and failure in passengers forecast for the East Coast mainline.

The week started with Government estimates of the effects of different types of ‘Brexit’ on the country including its regions. The North East was forecast to be most severely hit with a loss of 16% of GDP over the next 15 years. Unsurprisingly this report was not met with enthusiasm. As ever the focus was on shooting the messenger with Jacob Rees-Mogg and Ian Duncan Smith all declaring we should have not faith in these economic forecasting incompetents. It’s true that the macro forecasters didn’t cover themselves with glory at the time of the referendum. They may have shared a collective bias preferring the UK to vote remain. But are they any worse than the optimists? And what substantive grounds are there for optimism. We await the explanations. We know that there is no such animal as a good macro forecaster, only some who are less bad – and the consensus proves to be hard to beat. So to the Brexiteers – tell us why these forecasters are wrong and what better assumptions should be made?

The second controversial and in this case demonstrably inadequate forecast was for the number of passengers likely to use the East Coast mainline rail service operated by Stagecoach and Virgin. They had just been forced to renege on their government contract – the root cause was a failure to achieve the forecast increase in passenger numbers. No explanations have been forthcoming as to why the forecasts were so wrong. Some obvious possibilities come to mind – high forecasts were necessary to justify the bid: the decision was taken perhaps on the overall size of the bid and the numbers then massaged to justify the very substantial payments to the Government. The Governments Department of Transport would be disinclined to challenge them – after all, they were offering the Government more money than expected. Essentially the target numbers needed to sustain the investment became the forecast.

Two fallacies of forecasting are at the core of these stories: mixing targets, or wishful thinking, with forecasts and the lack of clear evaluation criteria of what constitutes a good forecast. Forecasts are supposed to provide an objective view of the future of the variables of interest, be that the economy or the passenger numbers. On these predictions one can plan ahead, or control existing plans. But for these numbers to be useful, they have to be as reliable as possible. What assumptions are behind these forecasts? What information is included? What is the objective of these forecasts, and therefore a good set of criteria to assess them? What political biases are likely interfering with an objective view. Answering these questions can be the basis of constructive improvement of the forecasts, rather than merely shooting the messenger.

 “This week’s forecast” is a blog series, which covers commentaries, by members of the Centre for Marketing Analytics and Forecasting, about topical and controversial forecasts in the news.


Back to News