Reading the Zew: Freude und Angst

Yesterday's Zew Survey gave a surprisingly dour reading about the current state of the German economy, but added a  surprisingly positive result about the likely trajectory of the economy over the coming six months in its central  Zew Indicator. How should one read it? Indeed, apart from providing an insight into the mind of the German finance industry, can we learn anything much about the likely trajectory of the German economy.  

In the end, I think we can, but only if we read the survey results with care. The monthly Zew survey presents the views of a relatively small sample of the German financial industry, this month  it had 272 participants. For the 'current views' survey, they are  asked to indicate if they think currently the economic situation is 'good' 'acceptable (normal), or 'bad'. In the event, 18.2% thought it was 'good', 68.8% thought it was 'normal' and 13% thought it was 'bad'. The final reading index is simply the % 'good' (18.2%) minus the % 'bad' (13%), leaving a reading of 5.2 (%).

Similarly, for the headline economic sentiment index (the Zew Indicator), respondents are asked whether they think the overall economic situation will improve, worsen, or remain the same over the coming six months. In the event, 56.8% expect things to improve, 34.6% expect little change, and 8.6% think things will get worse – so the indicator reads a net 48.2.

There are several things to notice about this. First, both questions are subjective: it may be that German financial professionals are constitutionally fairly glum about current circumstances, but basically optimistic that things will get better. And in fact, that's exactly what the survey's history suggests: since 2000, the average reading of the 'current situation' index has been minus 13.5%, whilst the average reading of the 6m economic sentiment index has been +17.9.

But actually, that's misleading: Germany's financial industry isn't routinely glum about Germany's current circumstances, so much as bi-polar, swinging regularly from profound angst to roaringly positive consensus, and back again. To put this in context, the modestly positive readings found in the last four months are quite literally unprecedented this century. The structure of the survey doesn't compel such a result, and this sort of bipolar assessment isn't repeated as dramatically in assessments of the Eurozone as a whole, or the US, or the UK.

There is a conclusion to be drawn: history suggests it is most unlikely that the Zew current conditions survey will survive for long in its relatively phlegmatic state. Rather, we must expect the mood to resolve itself into either deep pessimism or triumphant optimism in the coming months.

The Economic Sentiment (the Zew Index) is a different matter altogether. February's survey was the strongest since Dec 2009. 

Before we welcome this as a precursor of the much-anticipated recovery of Germany leading the Eurozone out of recession, there are three things to understand about the Zew Indicator as it stands:

1. Although it is far more volatile than the 'current conditions' index, it is not significantly bi-polar – a wide spread of views is regularly attained.
2. On balance, the industry is generally optimistic (and why not? Equities do tend to rise over time).
3. These sentiments have no predictive power about how the economy will be experienced in six months time. If it were, one would be able to find some relatively stable relationship between the Current Situation Index which would echo Zew Indicator. In fact, the two are radically different.    
4. In practice, the Zew Indicator by itself is not a good indicator of GDP growth or even direction.

But this raises an important issue of the relationship between the two indexes: clearly the importance attached to the Economic Sentiment index will be a function of how Current Conditions are experienced. A Zew Indicator of 100 (ie, unanimous belief that things will get better) means something quite different if the current conditions index is extremely miserable (say, minus 50) than if it is already quite cheerful (say +50).

One can capture this relativity by adjusting  'how things are going to look' by 'how things are now: ie, one adds the current view to the expectation, to get a clearer idea of where things are going. 

When we do that, we find: 

First, the volatility of the index is contained, and the seemingly rather excessive belief that things generally get better is muted, with the average value since 2000 retreating to +4.4 (rather than +17.9, unadjusted).

Second, Although the adjusted index shows sentiment improved, this improvement does not represent a fundamental breakout from the travails of the last three years that the unadjusted index describes. Rather, this looks like a repeat of the bounce achieved at this time last year.  

Third,  we also find that the adjusted index gains  in explanatory and even predictive power . In particular, it has proved a reasonable indicator of the direction and strength of GDP growth (though admittedly, it's not something I'd want to model from).    

Conclusions: Read properly, then, it seems that the German financial industry, as surveyed by the Zew Institute, does have a pretty keen idea of what's going on in the economy.  So what are they really telling us this month?

1. The bounce in economic expectations is less dramatic than it is initially stated, closely resembling what happened this time last year.
2. The bounce in economic expectations is not yet sufficiently pronounced to compel a significant acceleration in GDP growth.
3. The view that the current situation is modestly OK is historically unprecedented – one would expect it to resolve into either gloom or joyous optimism very shortly.
4. If the Zew is going to be an early indicator of a modest recovery in the German economy, the most crucial thing to watch over the coming months is the 'Current Situation' index, rather than the headline Zew Index.