Expert economists - the appaling record of prediction

Conrad

I see the OECD has had to revise upwards.


As Michael Gove said on Andrew Marr last Sunday, Economists are a sub-class of expert, certainly not of the ilk of a surgeon or physicist

martinsurrey

Conrad wrote: »

I see the OECD has had to revise upwards.


As Michael Gove said on Andrew Marr last Sunday, Economists are a sub-class of expert, certainly not of the ilk of a surgeon or physicist

and yet surgeons used to think that washing your hands before surgery was pointless, and physicists used to think that light traveled through a medium called aether.

all were wrong, but dealt with what they knew at the time.

ask your doctor to predict what a baby will die of, and most the time they'll get it wrong.

ask a physicist what the next breakthrough will be and most of the time they'll get it wrong.

no matter what the field, predicting the future is a pretty hard job!

TrickyTree83

martinsurrey wrote: »

no matter what the field, predicting the future is a pretty hard job!

I'd agree with the above but caveat it with this, that if you asked a physicist to predict something based upon known accepted physics that more often than not (much more often) the prediction will be more or less correct.

Economists appear to be the only science in which your predictive models that were wrong can be used again for your next prediction. If they were indicative, i.e. they showed growth albeit lower than previously expected, they would have more weight attached to their predictions. But to be predicting contraction when the actual result is growth is just wrong, it's not even indicative. It should be a real egg on their face moment, but it just seems to be brushed aside, patting them on the back for trying and telling them "better luck next time".

martinsurrey

TrickyTree83 wrote: »

I'd agree with the above but caveat it with this, that if you asked a physicist to predict something based upon known accepted physics that more often than not (much more often) the prediction will be more or less correct.

Ask an economist the effect of a well understood variable in a controlled environment (say the short term effect of a BOE rate change on the exchange rate with the $), and my bet is it would be pretty accurate, but we ask them to predict a once in a lifetime event in a world with a million variables, out for years.

https://en.wikipedia.org/wiki/Castle_Bravo

There is an example of the brightest physicists in the world getting the yield of an atomic bomb out by a massive margin, 3 times more powerful than expected as they were trying a new type.

TrickyTree83

martinsurrey wrote: »

Ask an economist the effect of a well understood variable in a controlled environment (say the short term effect of a BOE rate change on the exchange rate with the $), and my bet is it would be pretty accurate, but we ask them to predict a once in a lifetime event in a world with a million variables, out for years.

https://en.wikipedia.org/wiki/Castle_Bravo

There is an example of the brightest physicists in the world getting the yield of an atomic bomb out by a massive margin, 3 times more powerful than expected as they were trying a new type.

Economics and Physics are apples and oranges, you can make a prediction based on the laws of Physics and expect to be accurate, if you're not, clearly there's something you've overlooked and you can revise and learn, part of the scientific method. With economics you simply cannot do this for economic predictions.

Current economic forecasts make assumptions about who is in political power, the policies they undertake, economic and otherwise. If Putin decided to invade Latvia tomorrow do you think that would be included in the economic forecasts related to Latvia and the knock-on effect of increased military spending in Europe? Do they have indices for such events? I think that's unlikely since you're asking them to predict world events and their impact. It's an impossible task I agree, and so weight shouldn't be given to their work in the same way it currently is.

At best economic forecasts should be treated in the same way as the weather in this country. If the weather person says it's going to rain, take a brolly. If it doesn't rain you've lost nothing and if it does you're prepared.

martinsurrey

TrickyTree83 wrote: »

Economics and Physics are apples and oranges, you can make a prediction based on the laws of Physics and expect to be accurate, if you're not, clearly there's something you've overlooked and you can revise and learn, part of the scientific method. With economics you simply cannot do this for economic predictions.

Current economic forecasts make assumptions about who is in political power, the policies they undertake, economic and otherwise. If Putin decided to invade Latvia tomorrow do you think that would be included in the economic forecasts related to Latvia and the knock-on effect of increased military spending in Europe? Do they have indices for such events? I think that's unlikely since you're asking them to predict world events and their impact. It's an impossible task I agree, and so weight shouldn't be given to their work in the same way it currently is.

At best economic forecasts should be treated in the same way as the weather in this country. If the weather person says it's going to rain, take a brolly. If it doesn't rain you've lost nothing and if it does you're prepared.

I agree 100%, my comments were about comparing Economists and Physicists/Doctors and what a terrible idea that was, but even the premise that they are always right is flawed.

People in this thread also focus on the extreme predictors out of a massive sample, and then tar everyone with the same brush, they also compare their medium/long term forecasts to instant results, which is again a crazy idea.

Conrad

Martin the point is economists belong to a branch of sub experts, it's not called the dismal science without reason.

The current stable of predictions are based on pure guesswork inputs, that err on the side of pessimism.

Let's stick to the book rather than the crystal ball. The book says the economy is doing good, and massive corporations agree as they continue to invest.

wotsthat

The group 'Economists for Brexit' are saying everything is going to be great on just about every measure.

Those economists are all right aren't they?

daveyjp

Economics. The only exam where the answers, not the questions, change every year.

CLAPTON

wotsthat wrote: »

The group 'Economists for Brexit' are saying everything is going to be great on just about every measure.

Those economists are all right aren't they?

of course they are not :
first you are wrong as usual in saying that they claim everything will be great and secondly sensible people maintain a balanced views of the issues.