It is a difficult time. No one knows when the economic and financial crisis will end, leaving room to a rosier situation. Uncertainty on the economic future induces firms to postpone medium-long term decisions, e.g. to make an investment. If all firms behave this way, economic recovery becomes less likely.
Measuring political uncertainty. According to a recent piece on voxeu.org by Baker, Bloom and Davis, a peculiar feature of the current economic phase is that a sizeable amount of total economic uncertainty is due to economic policy uncertainty, i.e. regarding the decisions politicians and central bankers (do not) take. How can one measure this political uncertainty? In a nutshell, the idea is to count the daily number of Google News articles in which economic, political and uncertainty-related terms jointly appear. This raw figure is then divided by the total number of daily articles in which the word “today” appears, with the purpose of taking into account the exponentially increasing number of media outlets Google News has been indexing. The global political uncertainty index proposed by the three authors is a weighted average of various sub-indices: the political news component has a weight of one half, while the other sub-indices are the number of U.S. tax laws that expire each year, and the variability in inflation and government spending forecasts.
So. As one can gather from Figure 1 in the voxeu piece, the political uncertainty index has grown dramatically as of 2000; its peaks coincide with salient events such as 9/11, the second Iraq war, the collapse of Lehman Brothers and the political stalemate in the U.S. Congress on the debt ceiling.
The economic consequences of political uncertainty. Exploiting a statistical technique called VAR (one of the reasons why Chris Sims was recently awarded the Nobel Prize in Economics), Baker, Bloom e Davis show how an increase in political uncertainty does systematically foreshadow a decrease of industrial production and employment. In the lack of a proper experiment, one cannot safely conclude that this is evidence of a causal effect. Still, this is a relevant and interesting correlation, from a quantitative viewpoint as well: According to the authors’ calculations, “[…] restoring 2006 levels of policy uncertainty could increase industrial production by 4% and employment by 2.5 million jobs over 18 months.[…]”
What about Italy? Maybe there is no need to say it, but political uncertainty might be costly for other countries than the U.S., e.g. for Italy. In a recent piece for lavoce.info, I analyse to what extent during the last summer the difference between the spread on Italian long term government bonds (so called BTP) and Spanish ones (vis á vis the German Bunds) is in fact explained by politically relevant news on the Spanish and Italian press. An increase in the number of stories on the Italian Corriere della Sera regarding the budget law is significantly correlated with an increase in the differential spread, i.e. against Italy. By reading the articles, one can clearly check that the peaks in coverage occur when the budget approval process gets bumpy and slows down.
What if political uncertainty shrinks? I also provide some evidence that the difference between Italian and Spanish spreads significantly shrinks if the day before there are more articles on Corriere featuring the word “wiretappings”. One cannot conclusively prove anything with this simple data and analysis. Still, results are consistent with the hypothesis that financial markets appreciate Italian government bonds more when the probability of a political exit of the current prime minister is judged to be larger, maybe because of (additional) criminal prosecutions.