Thread by @benmarrow

Absolutely incredible paper presented at the @StiglerCenter's Political Economy of Finance conference in Chicago this weekend, by @NicolaLimodio.

Takeaway: fluctuations in the international price of silver can predict the probability of terrorist attacks!

(1/9)

How does this work? The Quran requires that muslims above a certain threshold of wealth give away 2.5% of wealth as charity ("zakat").

In Pakistan, the government automatically taxes citizens above this threshold at the 2.5% level.

(2/9)

Here's the key: the threshold ("nisab") is historically determined as the price of 612 grams of silver.

So each year, the Pakistani government looks at the current price of silver, and taxes everyone who falls above the corresponding threshold.

(3/9)

Individuals who fall *just* below the threshold one year due the chance price of silver suddenly avoid a 2.5% gvt tax. Those people then increase their private donations to charity (since they have more disposable income + islam calls for alms-giving).

(4/9)

Many such donations (whether intended or not) go to charities affiliated with terrorist groups. With this additional financing, these groups conduct more attacks:

↑silver price → ↑tax threshold → ↓taxes → ↑charity → ↑terrorist financing → ↑terrorist attacks!

(5/9)

Seem farfetched? Here's the key identification point: only *sunni* muslims have to pay to the tax. So, if this story is correct, the price of silver should predict sunni terrorist attacks in Pakistan but not shia ones...and only during Ramadan, when the tax is paid.

(6/9)

And that's what the data shows!

In graph below, the blue line shows the price of silver. The red line plots the *relative* probability of sunni terrorist attacks vs. shia ones.

(7/9)

If you're still not convinced, the paper has a litany of robustness checks, (not to mention a cool exercise that scrapes 2.5 million terrorist chat board messages from the dark web, and applies a NLP algorithm to interpret their content to look at terrorist demand)

(8/9)

All in all an incredible paper, and a real privilege to hear presented. Congrats to @NicolaLimodio.

Thanks also @gcaw for good discussion of paper afterwards. Lots of unanswered questions re: determinants of terrorism.

Read the full paper here: aeaweb.org/conference/201…

(9/9)