Heisenberg Report

Monday Night Read: One Chart And Some “Hard” Thoughts On “Soft” Data

Right, so if you have a fragile ego and are the type of person who is out looking for confirmatory evidence for your optimistic outlook for the US economy, you’re probably in a bad mood.

Why? Simple: all the data has been bad over the past couple of days.

Friday we got the first read on Q1 GDP and it was bad (although heading into the actual print it kinda felt like expectations had been set low enough that the market was able to take the number in stride). And then this morning, the Fed’s favorite econ indicator dipped (back) below target, spending came in light, and to top it off, the ISM and Markit prints were lackluster.

So heading into the Fed meeting (which, you’re reminded, is sure to be a non-event), there are very real questions about the extent to which the economy is on sound enough footing to warrant even cautiously optimistic projections and/or a continuation of the policy normalization trend.

Within that debate is the now ubiquitous “hard” versus “soft” data argument. No matter which side you fall on (i.e. whether you think the still relatively euphoric soft data is what matters of whether you think hard data “reality” is what counts), you should take a few minutes and consider the latest from former FX trader Richard Breslow, who was out this morning weighing in on what has become perhaps the most contentious question in the econ world.

But before we leave you in Breslow’s capable hands, do consider one chart which we think certainly seems to raise a number of questions about whether it makes sense for the ECB to be as far behind the Fed as they are in terms of normalizing policy:

Via Bloomberg

I’ve really been of two minds with respect to the hard versus soft data debate. On the one hand, it can be argued that soft data, like talk, can be cheap. It may or may not have some predictive value for the future. And we can all stipulate that economic forecasting is an inexact science. Hard data, on the other hand, is something you can touch and feel. So shouldn’t it be easy to decide which data set to follow? I used to think so.