Model answers…

2050. Not that far away eh? Alright it is still 38 years away but don’t worry, our banks and their economists are beavering away looking into their crystal balls so that we can start developing our future business strategies. A recent presentation that I attended concluded, perhaps unsurprisingly, that the top 10 economies in the world may not be same by the time we are halfway through the current century, and that countries that have low per capita income now will have the fastest growing economies of the future.

If you are of the opinion that the forecasting records of such experts are less than impressive, then you would no doubt have taken the elements of such an overview which could not be categorised as “bleedin’ obvious” with a sizeable dose of sodium chloride. Nonetheless many corporations lap these things up and focus their resources accordingly.

Obviously the forecasts above were dependent on a number of assumptions based on demographics, availability of resources and capital, and increasing levels of education, democracy and productivity. Oh and they used the history of economic development and applied it to their forecast model. The past is no guide to the future or those who forget history are condemned to relive it? Well you pays your money….

However one constant assertion particularly caught my attention. Although the assumptions might have been wide ranging, the model used was “extremely robust”. Oh really? I presume by this that they meant that it all added up and the calculations worked. I can’t think what else apart from the fact that it also produced pretty graphs and charts.

Having put together a number of forecasting models I know that however good you are at excel, and however many checks and balances you put in place, your output can only be as good as your inputs. If the assumptions are fatally flawed then the model, however robust it is, will be flawed as well. It is vital that everything is questioned and verified, and that lessons from previous forecasting processes are learned and acted on.

Actually when trying to visualise the very long term future, and trying to judge what changes might take place, I find it helps to cast your mind back the same number of years.  The noted music journalist and magazine publisher David Hepworth recently mused that it was 49 years since the first Rolling Stones single was released in 1963, and if you went back 49 years from that point Gavrilo Princip's trigger finger was about to unleash the horror that was the First World War. The inhabitants of 1914 would never have been able to envisage the swinging decade epitomised by Jagger, Richards et al.

To me that puts this whole long term forecasting lark into perspective. Anybody going back 38 years ago from today would have found themselves in 1974. Our intrepid forecasters from that time almost certainly would not have assumed a 2012 of persistent inflation, petrol price hikes, rising unemployment, world recession, and collapsing financial institutions. They might have assumed that we would have learnt from their experiences. The chances of the citizens of 2050 benefiting from us having learned the lessons of our current predicament will tell you more about the future than a “robust” forecasting model will ever do.