Here's an interesting theory courtesy of The Monkey Cage:
Contractions in economic outputs due to drought increase the likelihood of democratic reform while short-term weather-related jumps in output decrease that likelihood. That is the core finding of a new article in the American Economic Journal: Macroeconomics by Paul Burke and Andrew Leigh (ungated version here, h/t to Kevin Lewis). The effects are substantial. The authors estimate that a one-year recession in an autocratic nation that reduced GDP per capita growth by 6 percentage points increases that country's probability of undergoing significant democratic reform in the next year by 8 percentage points. The authors are careful to point out that these are short-term effects that do not necessarily tell us anything about long-term relationships between income and democracy. That is: shocks in economic outputs may determine the timing of regime change rather than whether a country eventually becomes (and stays) more democratic.
I wonder if you were to track agricultural output and (as much as they can be quantified) democratic openings and closings in the Arab world, you would find such a correlation. Could the current regression seen in Morocco be explained by good crops in the last year, contributing to decent GDP growth? Can you explain the opening of 2004-2006 by a poorly performing economy in 1999-2003 that demanded that a new management team be put in place (in the shape of the Nazif government)? Did Egypt's relatively strong performance (in terms of GDP growth) between 2005 and now make it easier for the regime to pull back from that opening?
Interesting questions all, but I would still look first to specific contingencies: electoral cycles, the policies of external actors, psychology of key regime actors, etc.