Google and the Recession | Hal Varian speaks at the GAAC conference
– posted October 20th, 2009 by Niamh Phelan No comments
Google rolled out the big guns on Thursday, opening with a talk by Hal Varian, Google’s Chief Economist.
Hal gained popularity with Web Analysts globally when he outlined in a New York Times article that “the sexy job in the next 10 years will be statisticians”.
Searching for insights in the economy
Hal’s presentation called “Predicting the Current Time” covered the insights Google found from their Search and Trends tools, insights related to economic recovery.
Hal said you can take any economic index and build a forecasting model, based on these values:
- the series in the last months (to account for current trends)
- the series in the same month last year (to account for seasonal variations)
- add in the trends in relevant Google search terms
Google’s research shows that in the past, the number of new unemployment claims peaks at the end of a recession. This is true in American history: almost every American recession has had a peak in initial unemployment claims immediately prior to the end of the recession.
The amount paid out in benefits peaks several months later. In the past this number peaked 4 to 6 months after the height of initial unemployment claims, but in more recent recessions this number has peaked 18 months later. This is a lagging indicator.
In America, the number of initial unemployment claims appears to have peaked already, and it is likely that their recession concluded over the Summer; however, unemployment will not peak for some months.
A search in Google Insights for ’social welfare’, restricted to Ireland, gives the following graph:
Google Insights 'Social Welfare'
This directly reflects the unemployment rates in Ireland, as measured by the CSO:
Unemployment Rates Ireland
The numbers on the graph reflect the number of searches for a particular term, relative to the total number of searches on Google over time. They don’t represent absolute search volume numbers, because the data is normalized and presented on a scale from 0-100.
Is Google showing us the light at the end of the tunnel?
Digg this post Submit to del.icio.us Submit to technorati
Categories Life outside iQ, Web analytics
