Not that iQ knows all that many economists you understand. But even given our limited frame of reference, it was abundantly clear when we first met Tim Harford earlier this year that he’s somewhat of an aberration in his field. For one thing he was, well… interesting. Cool even. And get this. He has a nickname. You know, like a boxer or a wrestler or something.
Tim “The Undercover Economist” Harford. Told you: Cool. Perhaps best known as the voice behind the FT’s Dear Economist column, Harford spends a lot of time writing about ‘the small stuff’ – micro-economics – and he must (surely?!) be the only economist in the world who runs a problem page answering readers’ everyday questions by way of economics.
What illuminating insights might such a man offer on the current financial climate then? iQ decided to find out… and risked a slap by asking him to predict the future…
iQ: I bet 2009 has left you spoilt for choice when it comes to writing about the economy, no?
Tim Harford: Well, quite, but at the FT there are a number of good writers on the subject of the financial crisis. We’ve got Martin Wolf on big picture economics, and Gillian Tett, who I think was really very prescient about the potential problems inherent in derivatives. I leave them to talk about the crisis.
I try to take a slightly different angle. So I write pieces about why economists didn’t forecast the crisis. And the basic answer is that economists didn’t forecast the crisis because they never really forecast anything, as forecasting is actually very difficult.
One piece I wrote looked at how individuals respond to losing their jobs; what actually happens to them, what the impact is on their long-term job prospects, how they behave. Do they behave in a rational way when looking for new jobs? The answer is probably not.
Another piece complained about why everyone talks about how ‘we’ behave in a recession. Because there is no ‘we’. You have to understand that although the media likes to give us the averages, the truth is that individual experiences in a recession are overwhelmingly different. Some people do extremely well, some do extremely badly; most do exactly the same in a recession as they do at any other time. So before you start saying ‘this is how we’re behaving in tough times’, you’ve got to understand that, actually, for a lot of people, these aren’t tough times at all. Until you recognise that you don’t really see what’s going on.
iQ: Interesting. What about technology and its part in the current economic crisis? Clearly the predictive systems, checks, and balances went wrong somewhere?
TH: What IT enabled was very complex modelling that wasn’t necessarily any better than simple modelling, it just made it harder to see the faults. I wouldn’t say IT took a central role in the crisis, but I would say that the ability of financial analysts to create very sophisticated products that they probably didn’t really understand – and certainly their bosses and the regulators didn’t understand – was enabled by IT. Like anything else,
technology can be misused.
iQ: Is such misuse happening a lot do you think?
TH: Well, it’s always a risk. One thing I would say is that this idea that the banks created these products that
they knew to be toxic and then passed them on to poor unsuspecting idiots who all went bankrupt, with the banks then making huge profits, is stretching the truth. This was the story going around for a long time but I think we’ve all started to realise that there’s something wrong with it. I.e. the banks didn’t make a huge profit – instead they were stuck with a lot of the toxic products that they themselves had created.
So I wouldn’t go for the deliberate deceit story. I think it’s much more likely that people made these assumptions that were wrong and that very few knew how badly wrong these assumptions were.
These products, these derivatives, behaved in very unexpected ways (especially) if you got just a small part wrong. The part I think they got wrong was in the correlations between all these underlying assets – whose performance may or may not have been correlated with one another.
The products were constructed in such a way that if you got the correlations even a little bit wrong, you went from a product that was (largely) risk-free to one that was incredibly risky. It was like saying, ‘If we toss a coin a thousand times we’re really very sure that we won’t get eight hundred heads.
Based on this we know that the product is completely safe.’ But it turns out that they were actually only “tossing the coin” five times; if you do that you certainly can get four heads, which is the same proportion (80%).
That was the kind of mistake that was being made. Put simply like that and it’s very easy to see the problem, but if the whole thing is buried in a pretty complex spreadsheet it’s not so easy to see. All you need to do is then bet $60 trillion that you are right and you’ve got a major problem.
iQ: We’ve had big recessions in the past and we’ll likely see them again in the future. Would you say these things are simply cyclical?
TH: Well, we do. But it’s different every time, it seems, and some troughs are deeper than others. I think we have yet to work out how bad the situation is this time. Some people were saying this is no big deal, but a few
voices on Wall Street were saying, “No actually this is the worst financial crisis the world has ever seen but noone has noticed yet”. More recently people have started saying that it’s not so bad and that they’ve “got away
with it”, but I think there are probably a few twists and turns in this story yet.
iQ: OK. Now for the big, loaded Faustian question. Predictions for 2010?
TH: The simple answer is I don’t know. I think that’s the honest answer and I’ve written several times about
why that’s so. Almost, in theory, we don’t know because it’s just too damned complicated. Look at the track record of forecasters: it’s always very bad. And it’s not just the economists; it’s the anthropologists, the sociologists, the political scientists. It doesn’t matter who you’re talking to or what you’re asking them to make predictions about; they’re always terrible. So what’s going to happen in 2010? I don’t know.
If an economist is doing his job right he should be a bit like a doctor. You don’t ask a doctor: “How long am I going to live? What’s going to kill me? Will it be heart failure or cancer?” You say: “I’ve got a problem. My leg hurts. Can you take a look, figure out what’s going on and perhaps suggest a solution?”
The same thing should be true of an economist. You should go to an economist and say: “I’ve got a problem I want to solve”, not ask them what’s going to happen in the future. Unfortunately many people do (the latter) and some economists encourage it, and that’s not very wise.
There’s this psychologist called Philip Tetlock who collected 80,000 forecasts from 200 or 300 experts in
all fields, including economists. He spoke to academics, journalists, diplomats, and think-tanks and grouped their forecasts in three ‘buckets’: ‘Get better’, ‘Get worse’, and ‘Stay the same’. The question is did those experts beat “the chimp” strategy of simply choosing ‘better’, ‘worse’ or ‘stay the same’ at random?’ And the answer is no, they didn’t.
There’s no group of experts who beat the chimp: young, old, male, female, different disciplines, left wing, right
wing – it doesn’t matter.
I think that tells us all we really need to know about the difficulty of predicting the future. I don’t regard that as an indictment of expertise because expertise is clearly valuable, but let’s not misuse it. It’s not useful for making forecasts in any field really and certainly not in any field of social sciences.
iQ: That being the case, where do we all go from here?
TH: Well we need to help prevent a further financial crisis and there are various measures that look sensible – things like getting the banks to hold more capital. We also need to build more transparency into the banking
system and to build in more testing of new products. What’s very common in all industries – the computer industry being a good example – is that people will (inevitably) come up with ideas that don’t work. The banking industry came up with derivatives and credit default swaps.
Credit default swaps are only about 10 years old, and were about $70 trillion at the time the credit crunch began andwe realised they weren’t working in the way we thought they would. So: it’s perfectly common to create new ideas and for them not to work; what’s not common is that it’s such a huge and important market at the time you discover that it isn’t working. As such, we need to find some way of testing financial products earlier and discovering early on whether they’re working or not.
Now, I don’t know how to do that, but it’s very important to try and figure out a way. That’s not something that’s being emphasised by most commentators. They seem to be looking at other things like bonuses and capital right now.
