This paper presents a method to improve the one-step-ahead forecasts of the Spanish unemployment monthly series. To do so, we use numerous potential explanatory variables extracted from searches in Google (Google Trends tool). Two different dimension reduction techniques are implemented (PCA and Forward Stepwise Selection) to decide how to combine the explanatory variables or which ones to use. The results of a recursive forecasting exercise reveal a statistically significant increase in predictive accuracy of 10–25%, depending on the dimension reduction method employed. A deep robustness analysis confirms these findings, as well as the relevance of using a large amount of Google queries together with a dimension reduction technique, when no prior information on which are the most informative queries is available.

This paper uses time series of job search queries from Google Trends to predict the unemployment in Spain. Within this framework, we study the effect of the so-called digital divide, by age and gender, from the predictions obtained with the Google Trends tool. Regarding males, our results evidence a digital divide effect in favor of the youngest unemployed. Conversely, the forecasts obtained for female and total unemployment clearly reject such effect. More interestingly, Google Trends queries turn out to be much better predictors for female than male unemployment, being this result robust to age groups. Additionally, the number of good predictors identified from the job search queries is also higher for women, suggesting that they are more likely to expand their job search through different queries.