If food prices increase ten-fold, it's because the store shelves are empty. Rationing will happen before the price goes that high. If so having food on hand will improve your diet and extend the rationing allowance over months. If you need to move/relocate because you ran out of food, presume that other people would have the same idea and that the refuge receiving areas probably won't be that welcoming (they will have shortages too).
In terms of odds, lets say that failures happen every 5 years (to put an upper limit on it ... the world has had single failures in 2003, 2008, 2011, and 2012). That's a 20% chance per year of something westerners notice and then forget about(*). The odds of two independent(**) failures in the same year in any given year is 1/25 (4%). That's something that many people won't have personally experienced although they probably know someone who has (cf. 1970s energy crisis, WWII rationing). The odds of three failures in any given year is 1/125 (0.08%); that's the stuff of history books (cf. Great Irish Famine, Maunder Minimum).
The odds of entirely avoiding a two-failure year over the next 30 years would be (1-0.04)^30=29%.
So those are the rough a priori probabilities which are too high for me to just dismiss and punt on the problem.
They are the reason I put this suggestion #2 on my list. The odds for screwing oneself up metabolically, etc. via lifestyle choices and becoming dependent on a borked health care system are higher ... so that's why that was #1. Both are useful because failure is rather critical ... compared to e.g. one's laptop refusing to boot or being unable to repair one's washing machine, say.
(*) What is hard to compute is how "tough" the personal impact is. If one's food budget is 10% of one's income it's going to be less than if it was 40%. Revolutions and civic unrest have started from single crop failures in countries where the average food budget is 40%.
A threefer could increase crop prices by 4-500%. Not even Americans would laugh that one off. Having lots of money in the bank is not going to fix the problem. The supermarket would become illiquid and inefficient. Not to mention the social consequences of being the person outbidding everybody else in the store.
(**) These odds presume independent events, but given that they're usually weather driven and sometimes pest driven, one should expect a finite level of covariance which would increase the odds. The global price setting and market system are in any case not independent.