Oxford's oldest student newspaper

Independent since 1920

Interview: Tim Harford

Why is this particular cup of coffee this price here, and is it different somewhere else? Why is it different somewhere else? What are the possible explanations for that? Life is full of all kinds of puzzles: such as why is there a rental market in dinner jackets when you can buy a dinner jacket and use it many times, but there isn’t a rental market in wedding dresses even though people only use wedding dresses once?”
 At times, I temporarily forget that it is I, the interviewer, who is supposed to be asking the questions. Yet this is typical of Tim Harford’s approach to economics and to life more generally: ask lots of questions, reflect on what you see, and think about the world around you. His weekly column in The Financial Times and bestselling book The Undercover Economist are based on this entire premise.

“The real world applications, they are the most interesting. They make us think rigorously but at the same time they make us face up to some complexity. There might be all kinds of psychological and institutional details that do not easily fit into the models. But thinking about those details, that’s all part of being a good economist.”

Harford, who studied PPE at Brasenose before doing a masters degree in Economics, believes that many recent approaches to the subject have paid too little attention to what is actually happening in the real world. He thinks that while there is “tremendous value in the maths”, there is a risk that we “crowd things out that do not easily lend themselves to being expressed mathematically”. The popularity of Economics has soared in recent years. Since 2007, for example, the uptake of Economics A-level has risen by about 50 per cent. Harford thinks that this can only be a good thing.

“It is a great subject and I think it is absolutely fascinating; the different mixture of topics that you have to wrestle with and the importance of the ideas and the subject matter.” However, Harford accepts that this was not part of any attempt on his part or anyone else’s to popularise economics. The rising interest in the subject is more of a “symptom of a bad thing, which is that we have had a catastrophic financial crisis. There is always going to be more interest in Economics – certainly in the media – as a result of that. People have realised that it matters.”

Given that the study of economics is assuming ever greater importance, I question him on the nature of statistics. By analysing variables using alternative methods, statistical analyses generate different numbers. Inflation, unemployment, and economic growth, for example, can all be measured using different methodological approaches and each approach tells a different story.

As Harford exemplifies, “Are median wages stagnating or not? It turns out that there are two perfectly good ways to ask the question and you get a different answer. If you look at everybody who is working, the median wage is stagnating. If you look at everybody who has had a job all year, the median wage is growing. So if you have a job, your wage is going up. But there are people quitting the labour force at the top with good jobs and people entering the labour force at the bottom with bad jobs and that means that the median wage isn’t going any where.”

Yet how are we to choose between alternative methods of measuring certain variables? Harford is quite clear on this issue: we do not need to make a choice because statistics “measure what they measure”. Rebutting my question, he argues, “To ask which statistic we should choose, that’s like asking which frame of your favourite film you should choose. Well no, you’ve got to watch the whole film.” His answer brings us back to the point about not being reductive in our approach to looking at the world. Statistics can be useful even though they “do not tell us everything”.

But the moment we think that they are telling us everything, that is when we encounter trouble. According to Harford, “We are in even more trouble when they start getting used as targets because then you immediately get distortions”.

The classic example of this phenomenon is ambulance waiting times. In the late 1990s, Tony Blair introduced a target: in life-threatening emergencies in urban areas, ambulances should arrive within eight minutes, 75 per cent of the time. “Even saying the target you realise that it is quite complicated,” comments Harford. “Immediately after that target was introduced, it started being hit for sure but the definition of life-threatening emergencies started changing. The definition of ambulance changed, so that suddenly paramedics would arrive on bikes. The ambulances were relocated from rural areas to urban areas to make it easier to hit the target in urban areas. People would just lie about the numbers and an awful lot of calls finished at seven minutes 59 seconds. And of course if you breach the eight minute target then you’re screwed because you’re no longer valid for the target so you might as well make them wait two hours.”

Harford continues, “You need to collect the data but the moment you turn the data into a target you are inviting all kinds of weird stuff.” This use of small real world examples to explain economic phenomena is typical of Harford’s approach.

However, Harford acknowledges that macroeconomics, which studies broader economic themes such as economic growth and unemployment, needs a “different approach”.

“Purely as a writer you have got to talk about macroeconomics in a different way. It is impossible to give a simple everyday example for a lot of 
everyday
 macro-
economic phenomena 
because the
whole point
is that they
 are systemic.
You cannot point
 to one cause because
 the moment you point to one cause there is something else going on behind your back which is connected via invisible strings to the other cause.” The nature of macroeconomics is such that there is significantly more controversy. Theories are inevitably less concrete, but Harford hopes that his writing on the subject is “still fun”.

One of the most significant aspects of macroeconomics is its attempt to forecast future events. Harford’s talk to the PPE Society following our interview was on ‘How to See Into the Future’. He talks of a new type of research which comes from psychologists rather than economists. This supposedly ground-breaking research characterises ‘super forecasters’ as those who work in a team, who have received training and, most importantly, those who adapt their forecasting technique in response to past errors.

Tentatively, I suggest to Harford that these characteristics give the impression of being somewhat superficial. Why are forecasters not already working in teams, partaking in training, and learning from their mistakes? He responds by suggesting that most forecasters do not face particularly strong incentives to attempt to produce accurate forecasts.

The chief economist at an investment bank is one example. “ Your job is to get the bank on the evening news and to sound authoritative.” Political pundits “may well just be cheering for their side in an argument,” whilst as an author trying to sell a book then “maybe you want to make a forecast which is eye-catching and controversial” rather than wholly accurate.
 Even independent think tanks such as the Institute for Fiscal Studies (IFS) and the Office for Budget Responsibility (OBR) have a relatively poor track record when it comes to forecasting. Indeed, not a single forecasting agency has correctly predicted a recession since 2007. Part of the problem, according to Harford, is that the “IFS for example is not usually graded on the accuracy of its forecasts, it is graded on other things: the detail, the independence of its analysis.” The principal job of independent forecasting agencies is, Harford concludes, to ensure that their forecasts are not “influenced by politicians trying to sell a particular idea”.

With no forecasting agency able to produce consistently accurate forecasts, I ask Harford why politicians, businesses and individuals commit so heavily to them.

“They’re very seductive, they’re very persuasive. They’re a great way of saying, ‘Oh well I don’t really know what’s going on in Syria; it’s all very complicated, but someone tells me that he believes that Bashar al-Assad will be out of office within six months.’ I feel like I have learnt something about Syria from an expert on Syria. I haven’t really but I feel like I have.”

The solution, according to Harford, is instead to engage in scenario-planning, “which is more about telling coherent, plausible stories about the future without making any claim that this story is definitely going to be correct.”

Reflecting on historical trends and real world problems as well as asking his own questions about what he sees appears to be Harford’s approach. It is an approach from which we all have much to learn.

Check out our other content

Most Popular Articles