Talking the economy down or up correlates not with fact

Robert Shiller would have a field day with this as it would certainly qualify for how and why the press sets the tone in socioeconomic matters regardless of fact. I woke up this morning to the following title from the German Reuters: “German Exports With Steepest Drop Since Crisis Year 2009”. In the report it is mentioned that exports dropped a whopping 5% month on month and that “economists expected a drop of 1,2%”.

I don’t know who these economists are and I have no idea what method they used to calculate what they expected but while I am not an economist I certainly did not expect a 1,2% drop. In fact whatever method the 1,2%-kind of economists used it did not include the Toll Index. Why I know this? Because if they did read my forecast 20 days ago they would have know that they have to expect a 5% drop. Here is my graph from 20 days ago:


The graph shows you the variation of the outbound heavy vehicles on German highways month on month for all Maut years (i.e. since 2007). It shows you a very robust month of August effect: it is the lowest point in the summer.  I checked the numbers to see what the drop represents in percentage points. It is as you can infer by inspection a drop from a value of 5.6 in July to 5.32 in August (went back and checked my data carefully). Now take out your abacus and do the math: 100x(5.32-5.6)/5.6 = 5. Voila!

The only thing that casts doubts in my mind on whether or not the Toll Index worked is that typically the first estimate of the official numbers are wrong and often by quite a bit.  I will be looking forward to the revisions. It could still be that the revisions will dance around the 5% for a few years only to come back and settle down on the 5% again…

Now if you did take a look at my post from 20 days ago you have noticed the difference in mood in my title (Toll Index August 2015 – crusing above and beyond) and in the Reuters’ title (which will now set the general tone for today and tomorrow in all the follow up reports of the rest of economic and financial press). So what accounts for this difference in mood?

In my view we had an (expected) drop and regardless of the percentage it drops to an exceptional level compared to other months of August in a time period which contains the Great Recession.  So my reporting was “cruising above and beyond” to express the fact that the 2015 string is way above the rest of the spaghetti lines. It is a drop but it is one which occurs from a high July to a high August. No big deal. Robert Shiller knows that economic reporting is not just reporting but it is a product which needs to be sold. So if you are the press the number is sellable only if it is tellable and hence  you tend to cherry pick that aspect of the variation which gives you a boiler room hurray or a doomsday drama. Today Reuters (and the rest of the press will) give us the latter. I say stay calm and carry on, it’s the data that counts.

PS: After all if you differentiate any polynomial often enough you will get  positive and negative numbers as you pick the coefficients along the way and eventually you will get zero. So if you don’t like the positive speed you may seek whether you have a negative acceleration and you can always choose between boiler room and doomsday if that’s what you want.

This entry was posted in comment, social, technology and tagged , , , , , , . Bookmark the permalink.

One Response to Talking the economy down or up correlates not with fact

  1. Pingback: Toll Index September 2015 – back to business as never before | signals in the noise

Comments are closed.