Brian Easton notes that journalists' misunderstandings about what makes a useful economic forecast results in entertaining commentary, and ignores the few forecasts that have been done rigorously

This is a re-post of an article originally published on pundit.co.nz. It is here with permission.

There are quality economic forecasts and there are entertaining ones.

Reviewing 2021, Stuff journalist Melanie Carroll pointed out that ‘Economists spend their days pondering what will happen [actually we don’t*], but last year they were taken by surprise by how strong inflation was, how well the economy withstood Covid, and Reserve Bank interest rate hikes.’

There follow a number of go-to economists – those whom the media regularly consult – explaining how they cocked up. Some failures arose because they were forecasting events over which economists have no special competence – like the course of a Covid virus. Others were economic events which the economists misjudged.

Carroll concludes that ‘In uncertain times it was better for businesses to scan the environment, read widely, and think about potential scenarios that affected their operations, rather than pin all their fortunes on the forecasts of one economist.’

I do not disagree with her conclusion but despite it, journalists will continue to go to economists for forecasts. They want short, pithy, certain, ones. The more competent the economist, the more caveats and admissions of uncertainty and so the less likely the economist is to be a go-to choice. It is a well-established research finding that the more certain a forecast is, the more likely it is to be wrong – as Carroll found.

If you want to see an economic forecast done properly, have a look at the Treasury Economic and Fiscal Updates, the last of which was published just before Christmas but was actually locked up in early November.

The Treasury is required by law to produce two a year (and three in an election year). That has the advantage that they are considered forecasts rather than instant responses to the latest statistic which will be subject to statistical noise and can be misleading if they are considered alone out of context of all the other data. Treasury also provides one of the most detailed forecasts; its economic outlook comes to some 20 pages and is in sufficient detail that if a good economist disagrees with it – say, over a particular assumption – then one can modify the Treasury forecast. Indeed, the Update itself often reports how a different assumption will change its figures.

To be clear, the Treasury predictions are not accurate. We don’t need Carroll to tell us that. The Treasury goes to a lot of trouble to explain how they got the previous one wrong (in this case the 2021 Budget update), including tables and graphs which compare them. (If you ask for them, as the OIA allows you to, the Treasury has more detailed comparisons of their forecasts showing their errors over a much longer period.)

Moreover, Treasury will avoid predicting things if they cannot. For instance, these forecasts mention the possibility of an Omicron outbreak but it would have been crazy to have included a scenario in the forecasts at that stage of the cycle.

What then is the point of Treasury making such resource-consuming forecasts, other than they are required by law? Traditionally their brief was to inform and support the Minister of Finance; in an open democracy that applies to all of us. The election update is specifically required so that the Opposition and electors are (almost) as informed as the Minister; not everyone uses the opportunity. The Minister then uses this and a variety of other information to inform his revenue, spending and borrowing plans.

Moreover the detail of the update is such that those who disagree with any assumption can modify the forecast. When a new piece of data comes in it will be checked against the forecast and if it is too far out of line, some adjustment will be made and the Minister informed. If a journalist asks him he will give a coherent response; an OIA can follow if anyone is dissatisfied.

If there is a major shock to the economy – such as the Omicron outbreak or the Evergrande Chinese property developer crashing – it is a relatively simple exercise to modify the Treasury forecast. My bet is that although the Omicron announcement was made last Monday, the Minister will have a coherent revision predicting the impact by this weekend.

So that is why I keep the latest Update next to my desk.

While I have written here about the Economic Update, the Fiscal Updates have another interest. Again there are specific forecasts so all the above applies. But there is also 50 (yes, fifty) pages of the risks to the fiscal forecasts, which provide an insight into the hosts of potential challenges that the government faces – not all of them, only those which impact directly on government revenue, spending, and borrowing, but there are a lot. Those who closely monitor the government find the details fascinating. Anyone who wants to make an informed comment about a government department should check to see what the risks section of the Fiscal Update says.

But that is not really what Carroll was writing about. The go-to’s forecasts are for your entertainment, not to inform you. Which is why the journalists keep reporting them.

* When Bob Solow said economists mainly think about sex, he may have been joking.

Brian Easton, an independent scholar, is an economist, social statistician, public policy analyst and historian. He was the Listener economic columnist from 1978 to 2014. This is a re-post of an article originally published on pundit.co.nz. It is here with permission.