That's not how it works. It's not rational to allocate resources to preparing for possible outcomes depending solely on the chance of that outcome happening. For example, there is maybe a ninety percent chance that I will get a splinter over the next year, but I don't allocate ninety percent of my resources preparing for this eventuality.
One rationally allocates resources so as to maximize expected utility; if an emergency has a very high negative utility, and an agent believes they can substantially minimize the utility loss by preparing thoroughly, it may work out for them to spend a fair amount of money preparing for a hypothetical emergency.