TED talks are great at giving you a brief overview of an area (or more often, one view of an area). So I usually like watching the non-economic talks. In contrast, the economics talks always leave me with more questions than insight. So it was with some trepidation that I watched a recent talk by Tim Harford available here.
I recommend it. The talk helps remind us of our own limitations and biases – and that is always a good thing.
These limitations are always useful to remember. When I came back to academia from the ACCC, I was asked by a colleague about merger analysis. How did the ACCC go about analysing the likely future effects of a merger? My reply was simple. We used a range of economic models that roughly described the industry. We used data and market inquiries to hone these models. We then used the alternative models to try and get likely patterns of behaviour. If the evidence and economics together was ambiguous, we worked out what information would separate the models and tried to get that information. There was no one ‘correct’ model and any theoretical results had to be compared and contrasted against the information from market inquiries and common sense.
My colleague was horrified by this reply. Apparently they had assumed that there was a ‘correct’ model that the ACCC could apply to get the ‘right answer’ for the competitive effects of a merger. My answer showed the limitations of economic analysis. I saw this as a benefit. My colleague had the opposite view. So I seem to be in Harford’s camp – and maybe the talk simply reinforced my own biases!