Political Forecasting: Anybody Can Do It; That Doesn’t Mean Everybody Should

I hadn’t intended to write on this topic, but my boss pulled me aside yesterday and said he’d been inundated with emails over the weekend – some asking about the controversy, and some asking if we could do the same type of forecasting!

In case you missed what I’m talking about, here’s a snippet (excerpt):

In tea leaves, economist sees slender Pakatan win

KUALA LUMPUR, Jan 11 — Pakatan Rakyat (PR) will notch a slim victory over the ruling Barisan Nasional (BN) in Election 2013, according to calculations by Bank Islam Malaysia’s chief economist Azrul Azwar Ahmad Tajudin.

In a report by The Straits Times of Singapore, Azrul’s computations found that BN was likely to win only between 97 and 107 of the 222 parliamentary seats — insufficient to reform the next administration.

But the result would also mean that PR will only gain a shaky hold over Parliament and far from the supermajority once enjoyed by its rival.

The outcome was considered the most probable during the banker’s presentation at the Regional Outlook Forum in Singapore yesterday, and took into account factors such as race and demographics.

Later, Azrul forecast that fallout would result from the PR win, with the stock market set to respond in “knee-jerk” fashion as well as an extended period of perceived instability.

He also did not rule out the possibility of “economic sabotage” by businesses and the civil service that are aligned with BN...

I’ve just read that Azrul has been suspended from duty by Bank Islam pending further inquiry, which was about the only course I would have expected the Bank to have taken.

I’ve met the man (crossed words with him too, at a recent panel discussion) – he’s obviously sincere in his beliefs and firm in his convictions. I won’t knock someone for standing by his principles. But I do have a few things to say about what’s occurred.

First and foremost, Azrul’s first duty is to his employer. Whether the Bank was aware he was in Singapore for the forum; whether he was there on an official or a personal capacity; by virtue of his position as Chief Economist, he represents Bank Islam at all times and at all places and should conduct himself appropriately.

This would be true of any management level officer in any company, never mind a GLC. To take a political stand, and especially one detrimental to the Bank’s image and standing, is solid grounds for dismissal.

It’s ok to be critical of government policies, as Azrul has been in the past. We’ve all done it at one time or another, because it can be done objectively and on the basis of learning and knowledge. Potentially embarrassing to his employer maybe, but still within bounds.

But by making an overtly political statement, Azrul crossed the invisible line between acceptable and unacceptable behaviour. Toeing the company line is part and parcel of corporate life, like it or not. And if you don’t like it, there’s always the option of leaving.

Secondly, there’s nothing in an economist’s training to allow him/her to make election forecasts, unless he/she specialises in political economy (and even then what you get is more along the lines of the political viability of government policies, rather than election outcomes). Azrul’s specialisation by the way is in finance.

Thirdly, few economists have formal training in forecasting – ok, I’ve let the cat out of the bag. This is most definitely true at undergraduate level, and mostly true even at postgraduate level. Most private sector economists do forecasts not because they have any particular competency in forecasting, but because they are expected to.

Same goes for many equity and fixed income analysts – they issue forecasts not because they know how, but because their clients demand it. As long as it sounds plausible and can be backed by data and logic, most get away with it, even if the forecasts bear little relation with what actually happens.

In my previous job, I used to resist making any kind of forecast, because I didn’t feel I knew enough to be credible or remotely accurate. Even now, whatever forecasts I do issue both officially and in this blog, are typically accompanied by range forecasts to provide an idea of the level of uncertainty (a salve to my conscience) and backed by statistical models and historical data. Frankly, I’m probably as bad at it as everybody else.

To provide an ostensibly “scientific” forecast on 1) something almost completely out of your area of competency and 2) is theoretically shaky and fraught with uncertainty, is at best unprofessional, at worst unprincipled and unethical.

As much as I admire Azrul for having the courage of his convictions, his sense of timing, appropriateness and choice of venue are atrocious. If he couldn’t stay silent on politics, he should have left Bank Islam.