Tag Archives: uncertainty

Krugman on Keynes and Uncertainty

Paul Krugman reviews Keynes: The Return of the Master by Robert Skidelsky in the Observer.  He writes:

…there’s an alternative interpretation of what Keynes was all about, one offered by Keynes himself in an article published in 1937, a year after The General Theory. Here, Keynes suggested that the core of his insight lay in the acknowledgement that there is uncertainty in the world – uncertainty that cannot be reduced to statistical probabilities, what the former US defence secretary Donald Rumsfeld called “unknown unknowns”. This irreducible uncertainty, he argued, lies behind panics and bouts of exuberance and primarily accounts for the instability of market economies.

In this book, Skidelsky puts himself in the camp of those who argue, in effect, that Keynes 1937, not Keynes 1936, is the man to listen to – that Keynesianism is, or should be, essentially about uncertainty and how it leads to economic instability. And from this he draws some radical conclusions.

Most strikingly, Skidelsky declares that the traditional division between microeconomics and macroeconomics, which is based on whether one focuses on individual markets or on the overall economy, is all wrong; macroeconomics should be defined as the field that studies those areas of economic life in which irreducible uncertainty, uncertainty that cannot be tamed with statistics, dominates. He goes so far as to call for a complete division of postgraduate studies: departments of macroeconomics should not even teach microeconomics, or vice versa, because macroeconomists must be protected “from the encroachment of the methods and habits of mind of microeconomics”.

How far should we be willing to follow Skidelsky in this? I think we must trust the biographer in his assessment of Keynes himself; Skidelsky argues persuasively that Keynes spent much of his life deeply focused upon, even obsessed with, the question of how one acts in the face of uncertainty, which is why Keynes 1937 comes closer to the essence of the great man’s own thinking.

That’s not the same thing, however, as saying that Keynes was right – even about his own contribution. Surely it’s possible to make the case for a less profound reconstruction of economics than Skidelsky advocates. I’d point out that behavioural economists, who drop the assumption of perfect rationality but don’t seem much concerned by the essential unknowability of the future, have done relatively well at making sense of this crisis; I’d also point out that current disputes over economic policy, above all about the usefulness of government spending to promote employment, seem to be primarily about Say’s Law – that is, Keynes 1936.

Uncertainty and climate change

Australian Economist, John Quiggin points out that uncertainty should increase intensity of climate change action.  He writes

…it’s a straightforward implication of standard economic analysis that the more uncertainty is the rate of climate change the stronger is the optimal policy response. That’s because, in the economic jargon, the damage function is convex. To explain this, think about the central IPCC projection of a 3.5 degrees increase in global mean temperature, which would imply significant but moderate economic damage (maybe a long-run loss of 5-10 per cent of GDP, depending on how you value ecosystem effects). In the most optimistic case, that might be totally wrong – there might be no warming and no damage. But precisely because this is a central projection it implies an equal probability that the warming will be 7 degrees, which would be utterly catastrophic. So, a calculation that takes account of uncertainty implies greater expected losses from inaction and therefore a stronger case for action. This is partly offset by the fact that we will learn more over time, so an optimal plan may involve an initial period where the reduction in emissions is slower, but there is an investment in capacity to reduce emissions quickly if the news is bad. This is why its important to get an emissions trading scheme in place, with details that can be adjusted later, rather than to argue too much about getting the short term parts of the policy exactly right.

Anyway, back to my main point. The huge scientific uncertainty about the cost of inaction has obscured a surprisingly strong economic consensus about the economic cost of stabilising global CO2 concentrations at the levels currently being debated by national governments, that is, in the range 450-550 ppm. The typical estimate of costs is 2 per cent of global income, plus or minus 2 per cent. There are no credible estimates above 5 per cent, and I don’t think any serious economist believes in a value below zero (that is, a claim that we could eliminate most CO2 emissions using only ‘no regrets’ policies).

For anyone who, like me, is confident that the expected costs of doing nothing about emissions, relative to stabilisation, are well above 5 per cent of global income that makes the basic choice an easy one.

Paul Saffo: Forecasting must embrace uncertainty

Futurist Paul Saffo recently gave a talk “Embracing Uncertainty – the secret to effective forecasting” at the Long Now foundation. The talk (mp3) and Stewart Brand’s summary are online on the Long Now Foundation website. The talk is similar to his article in Harvard Business Review Six Rules for Effective Forecasting (see also Podcast interview). His six rules are:

  1. Define a Cone of Uncertainty
  2. Look for the S Curve
  3. Embrace the Things That Don’t Fit
  4. Hold Strong Opinions Weakly
  5. Look Back Twice as Far as You Look Forward
  6. Know When Not to Make a Forecast

Saffo writes about forecasting:

The role of the forecaster in the real world is quite different from that of the mythical seer. Prediction is concerned with future certainty; forecasting looks at how hidden currents in the present signal possible changes in direction for companies, societies, or the world at large. Thus, the primary goal of forecasting is to identify the full range of possibilities, not a limited set of illusory certainties. Whether a specific forecast actually turns out to be accurate is only part of the picture—even a broken clock is right twice a day. Above all, the forecaster’s task is to map uncertainty, for in a world where our actions in the present influence the future, uncertainty is opportunity.

Unlike a prediction, a forecast must have a logic to it. That’s what lifts forecasting out of the dark realm of superstition. The forecaster must be able to articulate and defend that logic. Moreover, the consumer of the forecast must understand enough of the forecast process and logic to make an independent assessment of its quality—and to properly account for the opportunities and risks it presents. The wise consumer of a forecast is not a trusting bystander but a participant and, above all, a critic.