OptimISM or pessimISM?

It's not often, these days, that I find myself wishing I was back in an investment bank dealing room.  But it might have been fun to be there yesterday (Monday).  Here's why.

One of the most important tasks of a dealing room economist is to provide instant analysis of official data releases. As soon as the key data appear -- employment, inflation, GDP, for example -- the traders want to know how they should react.  As the "expert", you're expected within a minute to provide a quick rundown of the numbers themselves, including the key issue of how they relate to market expectations, and offer some guidance as to how the market might react.

For the most important numbers, this is going on simultaneously in every dealing room on the planet, and it leads to what a colleague of mine used to call the "8:31 effect".  Most of the key data in the US are released at 8:30 in the morning, so within a minute, all the economists have given their first take on what the numbers mean, and the traders are off to the races.

Now of course, after 8:31 the economists have time to dig deeper into the data.  Quite often the details convey a different message from the headline number, and when that happens, the "8:31 effect" can quickly unwind. Within an hour or so, the economists will have produced a written commentary on the data for traders and clients, and then it's time for a quick vente latte before starting to prepare for the next day's data.

So....on Monday of this week, the key scheduled data release in the US was the Institute of Supply Management (ISM) purchasing managers index (PMI) for May.  (It happens that this is one release that isn't published at 8:30, but let that pass).  The initial announcement was that the index had fallen from its April level, belying market hopes that the data would show the US manufacturing sector recovering from its weather-related difficulties in the first quarter of the year. The immediate impact, unsurprisingly, was that stocks took a bit of a hit.

Within the next hour or so, all of the economists at the big dealing houses would have put out a reasoned analysis of the data, giving their considered views of whether this was just a blip to be corrected within a month or two, or a sign that the US economy was facing greater headwinds than anticipated.  But wait!  A couple of hours after the initial release, the ISM popped back up on the screens to admit that actually, the index had risen in May, not fallen!  Cue a reversal in stocks, and a lot of economists hastily explaining to anyone who would listen why their firm conclusions of just a short while ago were now just dust in the wind.

And wait again! Shortly after the revised announcement, the ISM popped up again with an admission that it had misapplied its seasonal adjustment factors.  It now unveiled a third number, lower than the revised version but still showing an improvement over the April report.  Markets seem to have taken the ISM's word for it this time, because both the DJIA and S&P ended the trading day at new all time highs.

Nobody gets hurt and nobody dies when this sort of snafu occurs, but it's far from being a no harm, no foul kind of event. Huge sums of money ride on these data releases, and no doubt a few dealers who may have got caught on the wrong side will have called up the ISM (conveniently out of the way in Arizona) to demand assurances that this will never happen again.

And as for the poor economists, it must be hard to know where to turn when even the supposedly hard, real world data keep moving around in front of your eyes.  Maybe helps explain why so many academics like to shun the real world and stick to mathematical models.