Do economic forecasts depend on the nature of the economist?

The forecaster's personality seems to influence the forecast. While some are congenital optimists, others are pessimists. They are like the bulls and bears of economics-econometrics confluence

Recently, a very good economist friend of mine argued that we needed to move to quarter-on-quarter gross domestic product (GDP) forecasts. Currently, these forecasts are on a year-on-year basis. The latter, he said, missed the momentum part of GDP growth. Upon this, I thought to myself, “why not weekly then”?

Jokes aside, people always have, and always will, want to know what will happen in the future. Whether it is parrots picking out a card or an economist selecting the data, there’s not much difference really in the method. It all depends on what you trust.

So the question in the title is not such a frivolous one. It is exactly like “the glass is half-full” or “the glass is half-empty” statements.

NCAER became a pioneer in forecasting in 1998 under the able guidance of its head, Rakesh Mohan, and its chief economist, Shashanka Bhide. It was the first to develop its own forecasting model. All self-respecting economic organisations now feel compelled to have their own forecasts. I think there are about two dozens of them now. I am sure they have their own models which sometimes work and sometimes don’t. It doesn’t matter because there is no external accountability. 

Forecasts and predisposition

But models aside, I have always wondered if the personality of the forecaster influences the forecast because I know many of these forecasters. While some are congenital optimists, others are pessimists. They are, if you like, the bulls and bears of the confluence of economics and econometrics.

If we try, we can see the pattern in these forecasts. I invite you to draw up a list and see which economist is consistently optimistic and which one is consistently pessimistic. You will be surprised at how consistent they are.

The predisposition is determined by his or her non-economic beliefs. He or she will interpret the data accordingly. If you like Prime Minister Narendra Modi, it’s one thing. If not, it’s the other.

Bayes and Kahneman

Apart from prejudice, there are two factors at play here. Together they tend to dilute the accuracy and relevance of forecasts. This is a new post-2014 development. 

One factor is the basic Bayesian framework where beliefs play a decisive role. Here people look at signals about what's going on in the real world. Some of these signals are not clear and can be interpreted differently by different people.

What happens then is that people, instead of looking at the whole picture, interpret parts of it sequentially as the signals appear. It is only then that they form a full belief.

But because this is based on a series of interpretations, it can lead to beliefs that do not wholly square with facts. That is, beliefs are formed not on the basis of the signals but on the interpretation of those signals that vary from person to person.

The other factor is what I call the Kahnemanian Keynote or central theme. This is that you can’t predict the future accurately because the world is essentially non-linear.

I think this is what’s happened with India’s economic recovery. As my friend who wants quarter-to-quarter forecasts correctly put it, no one foresaw that India would delink mobility and the virus. In the end, we need to look at just one data point. And we can then make just one pukka forecast. 

The data point is that this is India. The forecast is that it will always spring a surprise.

Thus, GST collections have now come in higher than Rs 1 trillion for five straight months, the longest streak since GST started. Manufacturing PMI has been in expansion for seven straight months — flying in the face of predictions that an early recovery was driven by just pent-up demand and inventory re-stocking. And GDP growth has moved back to positive territory far faster than anyone expected. QED Twitter: tca_tca