The bumpy New Normal

By Felix Salmon
October 11, 2010
Mohamed El-Erian delivered this year's Per Jacobsson lecture at the IMF annual meetings, and was very clear that the international community has failed in its job over the past year or so:

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

Mohamed El-Erian delivered this year’s Per Jacobsson lecture at the IMF annual meetings, and was very clear that the international community has failed in its job over the past year or so:

The impressive degree of global coordination highlighted by the April 2009 G-20 meeting did not last long. It only took a few months for that moment of extraordinary collaboration to give way to solely domestic agendas.

The result of that, he says, is going to be ugly indeed:

Having won the war, industrial country societies are in the process of losing the peace. Indeed, absent some important mid-course corrections, industrial countries confront the prospects of low growth; high unemployment that is increasingly structural in nature; welfare losses, including a growing number of citizens falling through the large gaps created by overly stretched safety nets; and a rising risk of protectionism.

El-Erian is characteristically vague on exactly what those corrections should be, beyond saying, unhelpfully, that “structural reforms are key”. But the question’s probably moot in any case: countries are moving further apart, and tail risks are higher than ever.

El-Erian gave his speech on the same day that the ECB’s Lorenzo Bini Smaghi said that “if Greece restructures it would have a total collapse of the economy” — exactly the tail risk which is preventing Greece’s spreads from coming down to a sustainable level.

But if you want a clear visual of what the new world of higher tail risk looks like, you could do a lot worse than this chart, from the Bank of England’s latest inflation report, pointed to by El-Erian:


This is most emphatically not your typical bell curve, with the most likely outcome being represented by a peak in the middle. In fact, it’s inverted: the chances of inflation being outside the central 1.5% to 2.5% range are significantly higher than the chances of inflation falling within that sweet spot. And the UK’s central bank is, if anything, even more pessimistic:

The Committee judges that, given the scale of the risks in both directions, at both the two and three-year horizons there is only around a one-in-four chance that inflation will be within 0.5 percentage points of the 2% target.

This is a world which is out of the control of governments and central banks, where everybody is getting used to expecting the unexpected, and where uncertainty breeds a high degree of risk aversion even as monetary policy tries to push companies and investors to take ever-greater risks with their capital.

I’m inclined to agree with the message of El-Erian’s lecture: infighting between the world’s governments has failed the global economy, and we’re all going to be buffeted by unpleasant and unforeseeable consequences as a result. Fasten your seatbelts: the New Normal is going to be very bumpy.


We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see

Felix, I could turn any normal distribution into an “upside down” distribution as you did, simply by making each of the two central bars just half a standard deviation wide. You’ll get 19% in each of the central bars, then 31% in each of the side bars.

The distribution in question might or might not be normal, but I’m 100% confident that its true standard deviation is a lot closer than the 0.5% that is implied by the histogram construction.

Posted by TFF | Report as abusive

Felix, try the following link: ons/inflationreport/ir10aug5.ppt

Look at charts 5.6-5.10 (the one you show is 5.11). At least to a first approximation, these projections ARE normally distributed. They are simply normally distributed with a very wide spread.

Posted by TFF | Report as abusive