‘Googlenomics’ predicts sharp rise in US unemployment

It is no secret that I am quite fascinated by the the idea that social media data might be very useful as early/leading indicators of macroeconomic variables. Said in another way I think that social media activity can be seen as a form of prediction markets.

So recently I have been tracking what Google Trends is saying about the development in searches for different terms that might give an indication about whether we are heading for a recession in the US economy.

Lets start with the world ‘recession’.

Recession Google Trend

The picture is not dramatic and the Google searches for ‘Recession’ clearly is much lower than at the onset of the Great Recession in 2007-8. That said, there has recently been a relatively clear pick up in the ‘recession indicator’ that could indicate that ‘Google searchers’ are increasingly beginning to worry about the US economy entering a recession.

How about the labour market? This is Google searches for ‘Unemployment’.

Unemployment Google trends

This is much more alarming – there has been a very steep rise in Google searches for ‘Unemployment’. In fact the rise is more steep than it was in 2008. This certainly is an indication of a very sharp deterioration of US labour market conditions right now.

The question then is whether Google searches have any prediction power and here the evidence is quite clear that, that is indeed the case. At least that is the conclusion in a recent paper – The Predictive Power of Google Searches in Forecasting Unemployment – by Francesco D’Amuri and Juri Marcucci.

The evidence is in – the Fed should re-start QE rather than hike rates

Janet Yellen’s Federal Reserve have been extremely eager to say that inflation would soon rise due to the continued decline in unemployment and has essentially ignored all monetary and market indicators, which for a long time have indicated that monetary conditions should not be tightened as fast as the Fed has signaled.

That in my view is the main reason why US economic activity now is slowing significantly in and paradoxically that will now very likely push up unemployment. In fact if we trust the signals from Google searches then we are in for a significant deterioration in labour market conditions in the US very soon.

So while the Yellen-Fed seems to ignore monetary indicators at least the fact that unemployment might soon shoot up again should tell the Fed that it is time to dramatically change course.

In fact it now seems more likely that we will have a new round of Quantitative Easing in the next couple of quarters rather than more rate hikes. Or at least that is what the Fed should do to avoid another recession.

PS have a look at a couple of other Google searches as well: ‘jobs’, ‘loan+default’, ‘economic crisis’, ‘bear market’

PPS I seriously thought that Janet Yellen was well-aware of the dangers of repeating the mistakes of 1937. Apparently I was wrong.


If you want to hear me speak about these topics or other related topics don’t hesitate to contact my speaker agency Specialist Speakers – e-mail: or


Grexit, Germany and Googlenomics

The talk of Greece leaving the euro area – Grexit – is back. Will Grexit actually happen? I don’t know, but I do know that more and more people worry that it will in fact happen.

This is what Google Trends is telling us about Google searches for “Grexit“:


And guess what? While this is happening euro zone inflation expectations have collapsed. In fact this week 5-year German inflation expectations turned negative! This mean that the fixed income markets now expect German inflation to be negative for the next five years!

It is hard to find any better arguments for massive quantitative easing within a rule-based framework in the euro zone (with or without Greece). And this is how it should be done.

PS it has been argued recently that euro zone bond yields have declined because the markets are pricing in QE from the ECB. Well, if that is the case why is inflation expectations collapsing? After all investors should not expect monetary easing to led to lower inflation (in fact deflation) – should they?

PPS I do realise that the drop in oil prices play a role here, but the markets (forwards) do not forecast a drop in oil prices over the coming five years so oil prices cannot explain the deflationary expectations in Europe.

Google Trends: From Greek crisis to euro crisis

I have always found it fascinating how much information the internet is providing “real time” and  I believe that a lot of economic data can and should be “extracted” from data on internet activity. A good example is the development in google searches on the European debt crisis. Here Google Trends is an excellent tool.

We we input “Greek crisis” into Google Trends we get this timeline.

The graph is pretty clear – searches on the “Greek crisis” spikes in early 2010 and then start to ease off in May 2010. But then again from the early part of the Summer this year the “Greek crisis” start to take off again.

Then lets add one more search: “euro crisis”.

Here we compare the number of searches for “euro crisis” and “Greek crisis” respectively.

The picture is clear – there is a very high correlation between “Greek crisis” and “euro crisis”. However, there is more to tell. While “Greek crisis” searches is much higher the “euro crisis” searches in 2010 the picture has now changed.

Greek crisis becoming a euro crisis

Hence, since July the number of searches for “euro crisis” has outpaced the number of searches for “Greek crisis”. Zoom in on 2011 searches here.

Said in another way what Google Trends is telling us is that the Greek crisis has turned into a euro crisis. Is that a surprise? Maybe not, but it is an indication of the systemic risks involved in this terrible crisis.

%d bloggers like this: